EasyVisa Project

Context:

Business communities in the United States are facing high demand for human resources, but one of the constant challenges is identifying and attracting the right talent, which is perhaps the most important element in remaining competitive. Companies in the United States look for hard-working, talented, and qualified individuals both locally as well as abroad.

The Immigration and Nationality Act (INA) of the US permits foreign workers to come to the United States to work on either a temporary or permanent basis. The act also protects US workers against adverse impacts on their wages or working conditions by ensuring US employers' compliance with statutory requirements when they hire foreign workers to fill workforce shortages. The immigration programs are administered by the Office of Foreign Labor Certification (OFLC).

OFLC processes job certification applications for employers seeking to bring foreign workers into the United States and grants certifications in those cases where employers can demonstrate that there are not sufficient US workers available to perform the work at wages that meet or exceed the wage paid for the occupation in the area of intended employment.

Objective:

In FY 2016, the OFLC processed 775,979 employer applications for 1,699,957 positions for temporary and permanent labor certifications. This was a nine percent increase in the overall number of processed applications from the previous year. The process of reviewing every case is becoming a tedious task as the number of applicants is increasing every year.

The increasing number of applicants every year calls for a Machine Learning based solution that can help in shortlisting the candidates having higher chances of VISA approval. OFLC has hired your firm EasyVisa for data-driven solutions. You as a data scientist have to analyze the data provided and, with the help of a classification model:

• Facilitate the process of visa approvals.
• Recommend a suitable profile for the applicants for whom the visa should be certified or denied based on the drivers that significantly influence the case status.

Data Description

The data contains the different attributes of the employee and the employer. The detailed data dictionary is given below.

• case_id: ID of each visa application
• continent: Information of continent the employee
• education_of_employee: Information of education of the employee
• has_job_experience: Does the employee has any job experience? Y= Yes; N = No
• requires_job_training: Does the employee require any job training? Y = Yes; N = No
• no_of_employees: Number of employees in the employer's company
• yr_of_estab: Year in which the employer's company was established
• region_of_employment: Information of foreign worker's intended region of employment in the US.
• prevailing_wage: Average wage paid to similarly employed workers in a specific occupation in the area of intended employment. The purpose of the prevailing wage is to ensure that the foreign worker is not underpaid compared to other workers offering the same or similar service in the same area of employment.
• unit_of_wage: Unit of prevailing wage. Values include Hourly, Weekly, Monthly, and Yearly.
• full_time_position: Is the position of work full-time? Y = Full Time Position; N = Part Time Position
• case_status: Flag indicating if the Visa was certified or denied

Importing necessary libraries and data

#import sys
#!{sys.executable} -m pip install pandas-profiling

#Libraries to help with reading data and manipulating data
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt

# Libraries that support data visualization
import matplotlib.pyplot as plt
%matplotlib inline
import seaborn as sns

# Libraries to suppress warnings
import warnings 
warnings.filterwarnings("ignore")

sns.set()

# to split the data into train and test
from sklearn.model_selection import train_test_split

# to build logistic regression_model
from sklearn.linear_model import LogisticRegression

# to check model performance
from sklearn.metrics import mean_absolute_error, mean_squared_error, r2_score

# to build linear regression_model using statsmodels
import statsmodels.api as sm

from sklearn.tree import DecisionTreeClassifier
from sklearn import tree

# Removes the limit for the number of displayed columns
pd.set_option("display.max_columns", None)
# Sets the limit for the number of displayed rows
pd.set_option("display.max_rows", 200)


# To build model for prediction
import statsmodels.stats.api as sms
from statsmodels.stats.outliers_influence import variance_inflation_factor
import statsmodels.api as sm
from statsmodels.tools.tools import add_constant
from sklearn.linear_model import LogisticRegression


# To get diferent metric scores
from sklearn.metrics import (
    f1_score,
    accuracy_score,
    recall_score,
    precision_score,
    confusion_matrix,
    roc_auc_score,
    plot_confusion_matrix,
    precision_recall_curve,
    roc_curve,
    make_scorer,
)

from sklearn.model_selection import train_test_split
from sklearn.tree import DecisionTreeClassifier
from sklearn import tree
import sklearn.metrics as metrics

# To tune different models
from sklearn.model_selection import GridSearchCV

# Libraries to suppress warnings
import warnings 
warnings.filterwarnings("ignore")


from sklearn.ensemble import AdaBoostClassifier, GradientBoostingClassifier
#To install xgboost library use - !pip install xgboost 
from xgboost import XGBClassifier
from sklearn.ensemble import BaggingClassifier, RandomForestClassifier, StackingClassifier
from sklearn.linear_model import LogisticRegression

Data Overview

• Observations
• Sanity checks

url = "EasyVisa.csv"
visaData = pd.read_csv(url, index_col = 0)

#creditData = pd.read_csv("credit.csv")
visaData.head(10) #several missing values!

	continent	education_of_employee	has_job_experience	requires_job_training	no_of_employees	yr_of_estab	region_of_employment	prevailing_wage	unit_of_wage	full_time_position	case_status
case_id
EZYV01	Asia	High School	N	N	14513	2007	West	592.2029	Hour	Y	Denied
EZYV02	Asia	Master's	Y	N	2412	2002	Northeast	83425.6500	Year	Y	Certified
EZYV03	Asia	Bachelor's	N	Y	44444	2008	West	122996.8600	Year	Y	Denied
EZYV04	Asia	Bachelor's	N	N	98	1897	West	83434.0300	Year	Y	Denied
EZYV05	Africa	Master's	Y	N	1082	2005	South	149907.3900	Year	Y	Certified
EZYV06	Asia	Master's	Y	N	2339	2012	South	78252.1400	Year	Y	Certified
EZYV07	Asia	Bachelor's	N	N	4985	1994	South	53635.3900	Year	Y	Certified
EZYV08	North America	Bachelor's	Y	N	3035	1924	West	418.2298	Hour	Y	Denied
EZYV09	Asia	Bachelor's	N	N	4810	2012	Midwest	74362.1900	Year	Y	Certified
EZYV10	Europe	Doctorate	Y	N	2251	1995	South	67514.7600	Year	Y	Certified

Observations

1. no_of_employees and yr_of_estab are numrical variables. The rest are categorical Variables.
2. prevailing_wage is a float variable.
3. All the categorical variables should be converted to Categorical Variables.

#copying data to another data frame to avaoid changes in the original data
data_new = visaData.copy()

data_new.shape

(25480, 11)

Observations

1. There are 25480 rows and 11 columns

To check for the duplicates

# Selecting duplicate rows except first 
# occurrence based on all columns
duplicate = data_new[data_new.duplicated()]

duplicate

	continent	education_of_employee	has_job_experience	requires_job_training	no_of_employees	yr_of_estab	region_of_employment	prevailing_wage	unit_of_wage	full_time_position	case_status
case_id

Observations

1. There are no duplicate rows.

data_new.info()

<class 'pandas.core.frame.DataFrame'>
Index: 25480 entries, EZYV01 to EZYV25480
Data columns (total 11 columns):
 #   Column                 Non-Null Count  Dtype  
---  ------                 --------------  -----  
 0   continent              25480 non-null  object 
 1   education_of_employee  25480 non-null  object 
 2   has_job_experience     25480 non-null  object 
 3   requires_job_training  25480 non-null  object 
 4   no_of_employees        25480 non-null  int64  
 5   yr_of_estab            25480 non-null  int64  
 6   region_of_employment   25480 non-null  object 
 7   prevailing_wage        25480 non-null  float64
 8   unit_of_wage           25480 non-null  object 
 9   full_time_position     25480 non-null  object 
 10  case_status            25480 non-null  object 
dtypes: float64(1), int64(2), object(8)
memory usage: 2.3+ MB

Observations

1. All columns have 25480 observations indicating no missing values.
2. All the categorical variables should be converted to Categorical Variables.

data_new.describe().T

	count	mean	std	min	25%	50%	75%	max
no_of_employees	25480.0	5667.043210	22877.928848	-26.0000	1022.00	2109.00	3504.0000	602069.00
yr_of_estab	25480.0	1979.409929	42.366929	1800.0000	1976.00	1997.00	2005.0000	2016.00
prevailing_wage	25480.0	74455.814592	52815.942327	2.1367	34015.48	70308.21	107735.5125	319210.27

Observation

1. no_of_employees in some rows seems to be negative.
2. no_of_employees vary widely from -26 to 602069.
3. yr_of_est ranges widely between 1800 and 2016.
4. Also, Prevailing_wage is float variable ranging 2.137 to 319210.27. This also indicates that there are many outliers.

data_new.describe(exclude='number').T

	count	unique	top	freq
continent	25480	6	Asia	16861
education_of_employee	25480	4	Bachelor's	10234
has_job_experience	25480	2	Y	14802
requires_job_training	25480	2	N	22525
region_of_employment	25480	5	Northeast	7195
unit_of_wage	25480	4	Year	22962
full_time_position	25480	2	Y	22773
case_status	25480	2	Certified	17018

Observation

1. The variable continent has 6 unique variations.
2. The variables education_of_employee, unit_of_wage have 4 unique variations
3. Variable has_job_experience, requires_job_training, full_time_position, case_status have 2 variations.
4. The variable region_of_employee has 5 variations.

Missing value

 data_new.isna().sum()

continent                0
education_of_employee    0
has_job_experience       0
requires_job_training    0
no_of_employees          0
yr_of_estab              0
region_of_employment     0
prevailing_wage          0
unit_of_wage             0
full_time_position       0
case_status              0
dtype: int64

Observations

1. There are no missing values in the data provided.

category = ['continent', 'education_of_employee', 'has_job_experience','requires_job_training', 'region_of_employment',
           'unit_of_wage', 'full_time_position', 'case_status']

for column in category:
    print(data_new[column].value_counts())
    print('_'*40)

Asia             16861
Europe            3732
North America     3292
South America      852
Africa             551
Oceania            192
Name: continent, dtype: int64
________________________________________
Bachelor's     10234
Master's        9634
High School     3420
Doctorate       2192
Name: education_of_employee, dtype: int64
________________________________________
Y    14802
N    10678
Name: has_job_experience, dtype: int64
________________________________________
N    22525
Y     2955
Name: requires_job_training, dtype: int64
________________________________________
Northeast    7195
South        7017
West         6586
Midwest      4307
Island        375
Name: region_of_employment, dtype: int64
________________________________________
Year     22962
Hour      2157
Week       272
Month       89
Name: unit_of_wage, dtype: int64
________________________________________
Y    22773
N     2707
Name: full_time_position, dtype: int64
________________________________________
Certified    17018
Denied        8462
Name: case_status, dtype: int64
________________________________________

Observations

1. 66% of the cases are certified.
2. Most of the employees get yearly wages.
3. More than 80% of the employees are full-time

data_new.head()

	continent	education_of_employee	has_job_experience	requires_job_training	no_of_employees	yr_of_estab	region_of_employment	prevailing_wage	unit_of_wage	full_time_position	case_status
case_id
EZYV01	Asia	High School	N	N	14513	2007	West	592.2029	Hour	Y	Denied
EZYV02	Asia	Master's	Y	N	2412	2002	Northeast	83425.6500	Year	Y	Certified
EZYV03	Asia	Bachelor's	N	Y	44444	2008	West	122996.8600	Year	Y	Denied
EZYV04	Asia	Bachelor's	N	N	98	1897	West	83434.0300	Year	Y	Denied
EZYV05	Africa	Master's	Y	N	1082	2005	South	149907.3900	Year	Y	Certified

Check for unique values in the column

print(data_new.unit_of_wage.unique())

['Hour' 'Year' 'Week' 'Month']

print(data_new.region_of_employment.unique())

['West' 'Northeast' 'South' 'Midwest' 'Island']

print(data_new.education_of_employee.unique())

['High School' "Master's" "Bachelor's" 'Doctorate']

print(data_new.continent.unique())

['Asia' 'Africa' 'North America' 'Europe' 'South America' 'Oceania']

data_new.describe(include='all').T

	count	unique	top	freq	mean	std	min	25%	50%	75%	max
continent	25480	6	Asia	16861	NaN	NaN	NaN	NaN	NaN	NaN	NaN
education_of_employee	25480	4	Bachelor's	10234	NaN	NaN	NaN	NaN	NaN	NaN	NaN
has_job_experience	25480	2	Y	14802	NaN	NaN	NaN	NaN	NaN	NaN	NaN
requires_job_training	25480	2	N	22525	NaN	NaN	NaN	NaN	NaN	NaN	NaN
no_of_employees	25480.0	NaN	NaN	NaN	5667.04321	22877.928848	-26.0	1022.0	2109.0	3504.0	602069.0
yr_of_estab	25480.0	NaN	NaN	NaN	1979.409929	42.366929	1800.0	1976.0	1997.0	2005.0	2016.0
region_of_employment	25480	5	Northeast	7195	NaN	NaN	NaN	NaN	NaN	NaN	NaN
prevailing_wage	25480.0	NaN	NaN	NaN	74455.814592	52815.942327	2.1367	34015.48	70308.21	107735.5125	319210.27
unit_of_wage	25480	4	Year	22962	NaN	NaN	NaN	NaN	NaN	NaN	NaN
full_time_position	25480	2	Y	22773	NaN	NaN	NaN	NaN	NaN	NaN	NaN
case_status	25480	2	Certified	17018	NaN	NaN	NaN	NaN	NaN	NaN	NaN

Observation

1. The mean of no_of_employees is greater than 50% value, it is positive skewed.
2. yr_of_estab vary between 1800 and 2016.
3. prevailing_wage ranges from 2.137 and 219210.27.

data_new.isna().sum()

continent                0
education_of_employee    0
has_job_experience       0
requires_job_training    0
no_of_employees          0
yr_of_estab              0
region_of_employment     0
prevailing_wage          0
unit_of_wage             0
full_time_position       0
case_status              0
dtype: int64

Observation

1. There are no missing values.

data_new.isnull().sum()

continent                0
education_of_employee    0
has_job_experience       0
requires_job_training    0
no_of_employees          0
yr_of_estab              0
region_of_employment     0
prevailing_wage          0
unit_of_wage             0
full_time_position       0
case_status              0
dtype: int64

Convert the object variables to Categorical Variables.

['continent', 'education_of_employee', 'region_of_employment', 'unit_of_wage']

['continent', 'education_of_employee', 'region_of_employment', 'unit_of_wage']

data_new.continent = data_new.continent.astype('category')
data_new.education_of_employee = data_new.education_of_employee.astype('category')
data_new.region_of_employment = data_new.region_of_employment.astype('category')
data_new.unit_of_wage = data_new.unit_of_wage.astype('category')

data_new.has_job_experience = data_new.has_job_experience.astype('category')
data_new.requires_job_training = data_new.requires_job_training.astype('category')
data_new.full_time_position = data_new.full_time_position.astype('category')
data_new.case_status = data_new.case_status.astype('category')

data_new.has_job_experience.value_counts(dropna=False)

Y    14802
N    10678
Name: has_job_experience, dtype: int64

data_new.info()

<class 'pandas.core.frame.DataFrame'>
Index: 25480 entries, EZYV01 to EZYV25480
Data columns (total 11 columns):
 #   Column                 Non-Null Count  Dtype   
---  ------                 --------------  -----   
 0   continent              25480 non-null  category
 1   education_of_employee  25480 non-null  category
 2   has_job_experience     25480 non-null  category
 3   requires_job_training  25480 non-null  category
 4   no_of_employees        25480 non-null  int64   
 5   yr_of_estab            25480 non-null  int64   
 6   region_of_employment   25480 non-null  category
 7   prevailing_wage        25480 non-null  float64 
 8   unit_of_wage           25480 non-null  category
 9   full_time_position     25480 non-null  category
 10  case_status            25480 non-null  category
dtypes: category(8), float64(1), int64(2)
memory usage: 996.6+ KB

Observation

The object variables are converted to categorical variables.

Exploratory Data Analysis (EDA)

• EDA is an important part of any project involving data.
• It is important to investigate and understand the data better before building a model with it.
• A few questions have been mentioned below which will help you approach the analysis in the right manner and generate insights from the data.
• A thorough analysis of the data, in addition to the questions mentioned below, should be done.

Questions:

1. Those with higher education may want to travel abroad for a well-paid job. Does education play a role in Visa certification?

2. How does the visa status vary across different continents?

3. Experienced professionals might look abroad for opportunities to improve their lifestyles and career development. Does work experience influence visa status?

4. In the United States, employees are paid at different intervals. Which pay unit is most likely to be certified for a visa?

5. The US government has established a prevailing wage to protect local talent and foreign workers. How does the visa status change with the prevailing wage?

Exploratory Data Analysis

Let us explore Numerical Value first

#copying data to another data frame to avaoid changes in the original data
df = data_new.copy()

Observations on no_of_employees

sns.histplot(df.no_of_employees, kde = True)

<AxesSubplot:xlabel='no_of_employees', ylabel='Count'>

sns.boxplot(df.no_of_employees,orient = "h");

Observations.

1. The graph is positively skewed.
2. There are many outliers.
3. Most of the companies employee count is less than 20,000 Although there are companies where the employee count is 600000.

Observations on yr_of_estab

sns.histplot(df.yr_of_estab, kde = True)

<AxesSubplot:xlabel='yr_of_estab', ylabel='Count'>

sns.boxplot(df.yr_of_estab,orient = "h");

Observations

1. The graph is negatively skewed.
2. Most of the companies are established between 1980 and 2000.
3. There are many outliers.

Observations on prevailing_wage

sns.histplot(df.prevailing_wage, kde = True)

<AxesSubplot:xlabel='prevailing_wage', ylabel='Count'>

sns.boxplot(df.prevailing_wage,orient = "h");

Observations

1. The graph is positively skewed.
2. There are many outliers.

Lets explore the Categorical Variable.

def labeled_barplot(data, feature, perc=False, n=None):

    total = len(data[feature])  # length of the column
    count = data[feature].nunique()
    if n is None:
        plt.figure(figsize=(count + 1, 5))
    else:
        plt.figure(figsize=(n + 1, 5))

    plt.xticks(rotation=90, fontsize=15)
    ax = sns.countplot(
        data=data,
        x=feature,
        palette="Paired",
        order=data[feature].value_counts().index[:n].sort_values(),
    )

    for p in ax.patches:
        if perc == True:
            label = "{:.1f}%".format(
                100 * p.get_height() / total
            ) 
        else:
            label = p.get_height()  # count of each level of the category

        x = p.get_x() + p.get_width() / 2  # width of the plot
        y = p.get_height()  # height of the plot

        ax.annotate(
            label,
            (x, y),
            ha="center",
            va="center",
            size=12,
            xytext=(0, 5),
            textcoords="offset points",
        )  

    plt.show()  

Observations on continent

labeled_barplot(df, "continent", perc=True)

Observation

1. Most of the employees are from Asia.
2. Very less number of employees are from Oceania.

Observations on education_of_employee

labeled_barplot(df, "education_of_employee", perc=True)

Observation

1. 40.2% of the employees have Bachelors degree. 37.8% of the employees have Master's degree.
2. Very less number of employees(8.6%) of them have Doctor's degree.

Observation on has_job_experience

labeled_barplot(df, "has_job_experience", perc=True)

Observation

1. 58.1% of the applicants have job experience.

Observation on requires_job_training

labeled_barplot(df, "requires_job_training", perc=True)

Observation

1. 88.4% of the applicants donot require job training

Observations on region_of_employment

labeled_barplot(df, "region_of_employment", perc=True)

Observation.

1. Very less(1.5%) of the applicants are from Island.
2. Most of the employees are from Northeast, South and West regions.

Observations of unit_of_wage

labeled_barplot(df, "unit_of_wage", perc=True)

Observation

1. 90.1% of the employees receive year wages. 8.5% of the employees receive Hour wages.
2. 0.3% of the employees receive Month wages. 1.1% of the employees receive Week wages.

Observations on full_time_position

labeled_barplot(df, "full_time_position", perc=True)

Observation

1. 89.4% of the employees are in full_time_position. Rest are not.

Observations on case_status

labeled_barplot(df, "case_status", perc=True)

Observation

1. 66.8% of the cases are Certified.

Bivariate Analysis

Plotting the Bivariate Analysis to understand the interaction with each other

sns.heatmap(df.corr(),cmap="YlGnBu",vmin=-1,vmax=1, annot=True)

<AxesSubplot:>

Observation

1. Highest correlation from the above heatmap is between yr_of_estab and no_of_employees (0.018) which is very less.
2. It is important to note that correlation does not imply causation.
3. Least correlation is between no_of_employees and prevailing_wage

Bivariate Scatter Plots

sns.pairplot(df);

Observation

1. Highest correlation from the above heatmap is between yr_of_estab and no_of_employees (0.018) which is very less.
2. It is important to note that correlation does not imply causation.
3. Least correlation is between no_of_employees and prevailing_wage

Check continent by case_status

pd.crosstab(df.continent,df.case_status)

case_status	Certified	Denied
continent
Africa	397	154
Asia	11012	5849
Europe	2957	775
North America	2037	1255
Oceania	122	70
South America	493	359

sns.countplot(x="continent", hue="case_status", data=df, palette='Set1',saturation=50 );

Observation

1. Most of the applicants whose case status is certified are from Asia.
2. Most of the applicants whose case status is Denied are also from Asia.
3. Least rejections are for those aplicants from Oceania.

Check education_of_employee count by case_status

pd.crosstab(df.education_of_employee,df.case_status)

case_status	Certified	Denied
education_of_employee
Bachelor's	6367	3867
Doctorate	1912	280
High School	1164	2256
Master's	7575	2059

sns.countplot(x="education_of_employee", hue="case_status", data=df, palette='Set1',saturation=50 );

Observation

1. Most of the applicants whose case status is certified have Master's forllowed by candidates with Bachelor's degree.
2. Rejections are very less among Doctorate candidates.

Check has_job_experience by case_count

pd.crosstab(df.has_job_experience,df.case_status)

case_status	Certified	Denied
has_job_experience
N	5994	4684
Y	11024	3778

sns.countplot(x="has_job_experience", hue="case_status", data=df, palette='Set1',saturation=50 );

Observations

1. Most of the applicants whose case_status is certified have job experience.
2. There are more rejections among applicants whose case_status is denied.

check requires_job_training by case_count

pd.crosstab(df.requires_job_training,df.case_status)

case_status	Certified	Denied
requires_job_training
N	15012	7513
Y	2006	949

sns.countplot(x="requires_job_training", hue="case_status", data=df, palette='Set1',saturation=50 );

Observations

1. Most of the candidates whose case is certified are those who did not require job training.
2. Very less rejections are among those who required job training.

check region_of_employment by case_count

pd.crosstab(df.region_of_employment,df.case_status)

case_status	Certified	Denied
region_of_employment
Island	226	149
Midwest	3253	1054
Northeast	4526	2669
South	4913	2104
West	4100	2486

sns.countplot(x="region_of_employment", hue="case_status", data=df, palette='Set1',saturation=50 );

Observation

1. Most of the certified applicantions are from south region followed by Northeast and West.
2. Very less rejected applications are from Island.

check unit_of_wage by case_count

pd.crosstab(df.unit_of_wage,df.case_status)

case_status	Certified	Denied
unit_of_wage
Hour	747	1410
Month	55	34
Week	169	103
Year	16047	6915

sns.countplot(x="unit_of_wage", hue="case_status", data=df, palette='Set1',saturation=50 );

Observation

1. Most of the employees receive yearly wages.
2. The maximum cases are Certified among Yearly waged applicants.
3. Least rejections are among Monthly waged and weekly waged applicants.

Check full_time_position by case_status

pd.crosstab(df.full_time_position,df.case_status)

case_status	Certified	Denied
full_time_position
N	1855	852
Y	15163	7610

sns.countplot(x="full_time_position", hue="case_status", data=df, palette='Set1',saturation=50 );

Observation

1. Most of the certified applicants have full_time_position.
2. Very less rejected candidates have full_time_position.

no_of_employees Vs case_status

plt.figure(figsize=(15,5))

ax = sns.barplot(x='case_status', y='no_of_employees', data=df)

Observation

1. There are many Certified applicants than rejected.

prevailing_wage Vs case_status

plt.figure(figsize=(15,5))

ax = sns.barplot(x='case_status', y='prevailing_wage', data=df)

Observation

1. Certified applicants have more prevailing_wages than rejected applicants.

Relationship between no_of_employees and yr_of_estab

fig = plt.figure(figsize= (10,5))
ax = sns.regplot(x ='no_of_employees', y = df['yr_of_estab'], data = df)

Observation

1. The number of employees are more among the companies established between 1980 and 2000.
2. Very less number of employees belong to the companies established in 1800

Relationship between yr_of_estab and prevailing_wage

fig = plt.figure(figsize= (10,5))
ax = sns.regplot(x ='yr_of_estab', y = df['prevailing_wage'], data = df)

Observation

1. High Prevailing wages are among companies established between 1975 and 2000.
2. Less Prevailing wages are among companies established in 1800s.

Relationship between no_of_employees and prevailing_wage

fig = plt.figure(figsize= (10,5))
ax = sns.scatterplot(x ='no_of_employees', y = df['prevailing_wage'], data = df)

Observation

1. Most of the employees have prevailing_wages less than 150000.
2. Very less number of employees have higher prevailing_wages.

Correlation between no_of_employees, yr_of_estab and case_status

figure = plt.figure(figsize=(8,7))
sns.scatterplot(x="no_of_employees", y="yr_of_estab", data=df, hue='case_status', palette='tab10' )
plt.show()

Observation

1. Most of the cases certified are from the companies with employee count less than 200000 and also who belong to the companies established between 1950 and 2000.

Correlation between yr_of_estab, prevailing_wage and case_status

figure = plt.figure(figsize=(8,7))
sns.scatterplot(x="prevailing_wage", y="yr_of_estab", data=df, hue='case_status', palette='tab10' )
plt.show()

Observation

1. Most of the cases certified are employees who belong to the companies established between 1950 and 2000 and have prevailing_wages less than 150000.

Correlation between no_of_employees, prevailing_wage and case_status

figure = plt.figure(figsize=(8,7))
sns.scatterplot(x="prevailing_wage", y="no_of_employees", data=df, hue='case_status', palette='tab10' )
plt.show()

Observation

1. Most of the cases certified are employees who belong to the companies established between 1950 and 2000 and have prevailing_wages less than 150000.

Correlation between no_of_employees, continent and case_status

fig = plt.figure(figsize= (15,5))
ax = sns.barplot(x ='continent', y = 'no_of_employees', data = df, hue = 'case_status')

Observation

1. Less Rejections are among the applicants who belong to companies whose employee count is less and belong to the region Africa and Oceania.
2. Most of the cases are certified among the applicants from Africa followed by North America and Europe.

Correlation between no_of_employees, education_of_employee and case_status

fig = plt.figure(figsize= (15,5))
ax = sns.barplot(x ='education_of_employee', y = 'no_of_employees', data = df, hue = 'case_status')

Observation

1. Most of the cases certified are the applicants who belong to the company having more employee count and having Doctorate degree.
2. More rejections are among the applicants who have Doctorate degree.

Correlation between no_of_employees, has_job_experience and case_status

fig = plt.figure(figsize= (15,5))
ax = sns.barplot(x ='has_job_experience', y = 'no_of_employees', data = df, hue = 'case_status')

Observation

1. Equal number of cases are certified among the applicants with or without job experience. has_job_experience criteria is less dependent on the case_status.

Correlation between no_of_employees, requires_job_training and case_status

fig = plt.figure(figsize= (15,5))
ax = sns.barplot(x ='requires_job_training', y = 'no_of_employees', data = df, hue = 'case_status')

fig = plt.figure(figsize= (15,5))
ax = sns.pointplot(x ='requires_job_training', y = 'no_of_employees', data = df, hue = 'case_status')

Observation

1. Rejections doesnot depend if the candidates requires job training.
2. There are more cases certified among the candidates who donot require job training.

Correlation between no_of_employees, region_of_employment and case_status

fig = plt.figure(figsize= (15,5))
ax = sns.barplot(x ='region_of_employment', y = 'no_of_employees', data = df, hue = 'case_status')

fig = plt.figure(figsize= (15,5))
ax = sns.lineplot(x ='region_of_employment', y = 'no_of_employees', data = df, hue = 'case_status')

Observation

1. More cases are certified among the applicants who belong to the companies with more employee count and belong to Island.
2. Least rejections are among the applicants from Midwest. Maximum certified cases are among the applicants from Island.

Correlation between no_of_employees, unit_of_wage and case_status

fig = plt.figure(figsize= (15,5))
ax = sns.barplot(x ='unit_of_wage', y = 'no_of_employees', data = df, hue = 'case_status')

fig = plt.figure(figsize= (15,5))
ax = sns.pointplot(x ='unit_of_wage', y = 'no_of_employees', data = df, hue = 'case_status')

Observations

1. More applicants are certified who belong to the companies whose employee count is high and receive weekly wages.
2. More applicants are rejected among those who receive yearly wages.

Correlation between no_of_employees, full_time_position and case_status

fig = plt.figure(figsize= (15,5))
ax = sns.barplot(x ='full_time_position', y = 'no_of_employees', data = df, hue = 'case_status')

Observation

1. Less rejections are seen among the employees who have full_time_position and belong to the companies with more employee count.
2. More case are certified among the employees who do not have full_time_position.

### distribution of plot across each case_status - violin plot

ax = sns.violinplot(x =df.case_status, y = df['yr_of_estab'])

Observation

1. Most of the cases certified are among the employees who belong to the companies established between 1950 and 2000.
2. Very less cases are certified among the employees who belong to the companies established in 1800s.

ax = sns.violinplot(x =df.case_status, y = df['no_of_employees'])

Observation

There are many Certified applicants than rejected.

1. Those with higher education may want to travel abroad for a well-paid job. Does education play a role in Visa certification?

labeled_barplot(df, "education_of_employee", perc=True)

sns.countplot(x="education_of_employee", hue="case_status", data=df, palette='Set1',saturation=50 );

Observation

1. In the given dataset, there are more applicants with Bachelor's degree followed by Master's degree.
2. More cases are certified among the employees who have Master's degree followed by Bachelor's degree.
3. Least rejections are among the employees who have Doctorate degree.

2. How does the visa status vary across different continents?

sns.countplot(x="continent", hue="case_status", data=df, palette='Set1',saturation=50 );

Observation

1. More cases are certified among the employees who belong to Asia.
2. Least cases are certified among the employees who belong to Oceania.
3. Less rejections are made among the employees who belong to Africa followed by South America.

fig = plt.figure(figsize= (15,5))
ax = sns.barplot(x ='continent', y = 'no_of_employees', data = df, hue = 'case_status')

Observation

1. Less Rejections are among the employees from Africa and Oceania.
2. Most of the cases are certified among the applicants from Africa followed by North America and Europe.

3. Experienced professionals might look abroad for opportunities to improve their lifestyles and career development. Does work experience influence visa status?

labeled_barplot(df, "has_job_experience", perc=True)

sns.countplot(x="has_job_experience", hue="case_status", data=df, palette='Set1',saturation=50 );

Observation

1. From the given dataset, there are more employees who have job experience.
2. More cases are certified among the employees who have job exoerience.
3. More rejections are among the employees who donot have job experience.

fig = plt.figure(figsize= (15,5))
ax = sns.barplot(x ='has_job_experience', y = 'no_of_employees', data = df, hue = 'case_status')

Observation

Equal number of cases are certified among the applicants with or without job experience. has_job_experience criteria is less dependent on the case_status.

4. In the United States, employees are paid at different intervals. Which pay unit is most likely to be certified for a visa?

sns.countplot(x="unit_of_wage", hue="case_status", data=df, palette='Set1',saturation=50 );

Observation

1. Most of the employees receive yearly wages.
2. The maximum cases are Certified among Yearly waged applicants.
3. Least rejections are among Monthly waged and weekly waged applicants.

5. The US government has established a prevailing wage to protect local talent and foreign workers. How does the visa status change with the prevailing wage?

plt.figure(figsize=(15,5))

ax = sns.barplot(x='case_status', y='prevailing_wage', data=df)

Observation

Certified applicants have more prevailing_wages than rejected applicants.

Insights

1. Most of the cases certified and also rejected are among the applicants from Asia. Less rejections are from Oceania.
2. There are more certified applicants from Africa, followed by North America and Europe.
3. 40.2% of the employees have Bachelors degree. 37.8% of the employees have Master's degree. Very less number of employees(8.6%) of them have Doctor's degree.
4. Most of the applicants whose case status is certified have Master's forllowed by candidates with Bachelor's degree. Rejections are very less among Doctorate candidates.
5. 58.1% of the applicants have job experience. has_job_experience criteria is less dependent on the case_status.
6. 88.4% of the applicants donot require job training. Most of the candidates whose case is certified are those who did not require job training.
7. Most of the employees are from Northeast, South and West regions. A small number(1.5%) of the applicants are from Island. Most of the certified applicantions are from south region followed by Northeast and West. Very less rejected applications are from Island. Least rejections are among the applicants from Midwest.
8. 90.1% of the employees receive year wages. 8.5% of the employees receive Hour wages. 0.3% of the employees receive Month wages. 1.1% of the employees receive Week wages. The maximum cases are Certified among Yearly waged applicants. Least rejections are among Monthly waged and weekly waged applicants.
9. 89.4% of the employees are in full_time_position. Most of the certified applicants have full_time_position.
10. 66.8% of the cases are Certified. Most of the cases certified are among the employees who belong to the companies established between 1950 and 2000.
11. Certified applicants have more prevailing_wages than rejected applicants. High Prevailing wages are among companies established between 1975 and 2000 and employees having prevailing_wages less than 150000 .

Data Preprocessing

• Missing value treatment (if needed)
• Feature engineering
• Outlier detection and treatment (if needed)
• Preparing data for modeling
• Any other preprocessing steps (if needed)

# let's create a copy of the data
df1 = df.copy()

np.random.seed(1)

df1.sample(n=10)#Return a random sample of 10 rows from the dataframe 'data'

	continent	education_of_employee	has_job_experience	requires_job_training	no_of_employees	yr_of_estab	region_of_employment	prevailing_wage	unit_of_wage	full_time_position	case_status
case_id
EZYV17640	Asia	Bachelor's	Y	N	567	1992	Midwest	26842.9100	Year	Y	Certified
EZYV23952	Oceania	Bachelor's	N	N	619	1938	Midwest	66419.9800	Year	Y	Certified
EZYV8626	Asia	Master's	N	N	2635	2005	South	887.2921	Hour	Y	Certified
EZYV20207	Asia	Bachelor's	Y	Y	3184	1986	Northeast	49435.8000	Year	Y	Certified
EZYV7472	Europe	Bachelor's	Y	N	4681	1928	West	49865.1900	Year	Y	Denied
EZYV3434	Asia	Bachelor's	Y	N	222	1989	South	813.7261	Hour	Y	Certified
EZYV24441	Europe	High School	N	Y	3278	1994	South	204948.3900	Year	Y	Denied
EZYV12105	Asia	Master's	Y	N	1359	1997	West	202237.0400	Year	N	Certified
EZYV15657	Asia	Bachelor's	N	N	2081	2003	West	111713.0200	Year	Y	Denied
EZYV23111	North America	Bachelor's	Y	N	854	1998	Northeast	444.8257	Hour	Y	Denied

Missing Value Treatment

Check for missing value

df1.isnull().sum().sort_values(ascending = False) 
#Return the count of missing values column-wise and sort them in descending order

continent                0
education_of_employee    0
has_job_experience       0
requires_job_training    0
no_of_employees          0
yr_of_estab              0
region_of_employment     0
prevailing_wage          0
unit_of_wage             0
full_time_position       0
case_status              0
dtype: int64

Observation

There are no missing values. Hence, there is no need for missing value treatment.

Duplicate Rows Check

# Selecting duplicate rows except first 
# occurrence based on all columns
duplicate_new = df1[df1.duplicated()]

duplicate_new

	continent	education_of_employee	has_job_experience	requires_job_training	no_of_employees	yr_of_estab	region_of_employment	prevailing_wage	unit_of_wage	full_time_position	case_status
case_id

Observations

There are no duplicate rows.

Feature Engineering

df1

	continent	education_of_employee	has_job_experience	requires_job_training	no_of_employees	yr_of_estab	region_of_employment	prevailing_wage	unit_of_wage	full_time_position	case_status
case_id
EZYV01	Asia	High School	N	N	14513	2007	West	592.2029	Hour	Y	Denied
EZYV02	Asia	Master's	Y	N	2412	2002	Northeast	83425.6500	Year	Y	Certified
EZYV03	Asia	Bachelor's	N	Y	44444	2008	West	122996.8600	Year	Y	Denied
EZYV04	Asia	Bachelor's	N	N	98	1897	West	83434.0300	Year	Y	Denied
EZYV05	Africa	Master's	Y	N	1082	2005	South	149907.3900	Year	Y	Certified
...	...	...	...	...	...	...	...	...	...	...	...
EZYV25476	Asia	Bachelor's	Y	Y	2601	2008	South	77092.5700	Year	Y	Certified
EZYV25477	Asia	High School	Y	N	3274	2006	Northeast	279174.7900	Year	Y	Certified
EZYV25478	Asia	Master's	Y	N	1121	1910	South	146298.8500	Year	N	Certified
EZYV25479	Asia	Master's	Y	Y	1918	1887	West	86154.7700	Year	Y	Certified
EZYV25480	Asia	Bachelor's	Y	N	3195	1960	Midwest	70876.9100	Year	Y	Certified

25480 rows × 11 columns

# Lets look at the statistical summary of the data
df1.describe(include="all").T

	count	unique	top	freq	mean	std	min	25%	50%	75%	max
continent	25480	6	Asia	16861	NaN	NaN	NaN	NaN	NaN	NaN	NaN
education_of_employee	25480	4	Bachelor's	10234	NaN	NaN	NaN	NaN	NaN	NaN	NaN
has_job_experience	25480	2	Y	14802	NaN	NaN	NaN	NaN	NaN	NaN	NaN
requires_job_training	25480	2	N	22525	NaN	NaN	NaN	NaN	NaN	NaN	NaN
no_of_employees	25480.0	NaN	NaN	NaN	5667.04321	22877.928848	-26.0	1022.0	2109.0	3504.0	602069.0
yr_of_estab	25480.0	NaN	NaN	NaN	1979.409929	42.366929	1800.0	1976.0	1997.0	2005.0	2016.0
region_of_employment	25480	5	Northeast	7195	NaN	NaN	NaN	NaN	NaN	NaN	NaN
prevailing_wage	25480.0	NaN	NaN	NaN	74455.814592	52815.942327	2.1367	34015.48	70308.21	107735.5125	319210.27
unit_of_wage	25480	4	Year	22962	NaN	NaN	NaN	NaN	NaN	NaN	NaN
full_time_position	25480	2	Y	22773	NaN	NaN	NaN	NaN	NaN	NaN	NaN
case_status	25480	2	Certified	17018	NaN	NaN	NaN	NaN	NaN	NaN	NaN

category = ['continent', 'education_of_employee', 'has_job_experience','requires_job_training', 'region_of_employment',
           'unit_of_wage', 'full_time_position', 'case_status']

for column in category:
    print(df1[column].value_counts())
    print('_'*40)

Asia             16861
Europe            3732
North America     3292
South America      852
Africa             551
Oceania            192
Name: continent, dtype: int64
________________________________________
Bachelor's     10234
Master's        9634
High School     3420
Doctorate       2192
Name: education_of_employee, dtype: int64
________________________________________
Y    14802
N    10678
Name: has_job_experience, dtype: int64
________________________________________
N    22525
Y     2955
Name: requires_job_training, dtype: int64
________________________________________
Northeast    7195
South        7017
West         6586
Midwest      4307
Island        375
Name: region_of_employment, dtype: int64
________________________________________
Year     22962
Hour      2157
Week       272
Month       89
Name: unit_of_wage, dtype: int64
________________________________________
Y    22773
N     2707
Name: full_time_position, dtype: int64
________________________________________
Certified    17018
Denied        8462
Name: case_status, dtype: int64
________________________________________

df1.info()

<class 'pandas.core.frame.DataFrame'>
Index: 25480 entries, EZYV01 to EZYV25480
Data columns (total 11 columns):
 #   Column                 Non-Null Count  Dtype   
---  ------                 --------------  -----   
 0   continent              25480 non-null  category
 1   education_of_employee  25480 non-null  category
 2   has_job_experience     25480 non-null  category
 3   requires_job_training  25480 non-null  category
 4   no_of_employees        25480 non-null  int64   
 5   yr_of_estab            25480 non-null  int64   
 6   region_of_employment   25480 non-null  category
 7   prevailing_wage        25480 non-null  float64 
 8   unit_of_wage           25480 non-null  category
 9   full_time_position     25480 non-null  category
 10  case_status            25480 non-null  category
dtypes: category(8), float64(1), int64(2)
memory usage: 2.0+ MB

Convert the negative values to positive using Floor Function.

Since, the no_of_employees column containe negative values. This number cannot be negative. Hence, converting them to positive values using abs function.

df1 = df.copy()

df1.info()

<class 'pandas.core.frame.DataFrame'>
Index: 25480 entries, EZYV01 to EZYV25480
Data columns (total 11 columns):
 #   Column                 Non-Null Count  Dtype   
---  ------                 --------------  -----   
 0   continent              25480 non-null  category
 1   education_of_employee  25480 non-null  category
 2   has_job_experience     25480 non-null  category
 3   requires_job_training  25480 non-null  category
 4   no_of_employees        25480 non-null  int64   
 5   yr_of_estab            25480 non-null  int64   
 6   region_of_employment   25480 non-null  category
 7   prevailing_wage        25480 non-null  float64 
 8   unit_of_wage           25480 non-null  category
 9   full_time_position     25480 non-null  category
 10  case_status            25480 non-null  category
dtypes: category(8), float64(1), int64(2)
memory usage: 2.0+ MB

def use_floor_function(num):
    if num < 0:
        num = abs(num)
    return num

neg_var_col = ['no_of_employees']

for col_name in neg_var_col:
    df1[col_name] = df1[col_name].apply(use_floor_function)
    

df1.no_of_employees

case_id
EZYV01       14513
EZYV02        2412
EZYV03       44444
EZYV04          98
EZYV05        1082
             ...  
EZYV25476     2601
EZYV25477     3274
EZYV25478     1121
EZYV25479     1918
EZYV25480     3195
Name: no_of_employees, Length: 25480, dtype: int64

# Lets look at the statistical summary of the data
df1.describe(include="all").T

	count	unique	top	freq	mean	std	min	25%	50%	75%	max
continent	25480	6	Asia	16861	NaN	NaN	NaN	NaN	NaN	NaN	NaN
education_of_employee	25480	4	Bachelor's	10234	NaN	NaN	NaN	NaN	NaN	NaN	NaN
has_job_experience	25480	2	Y	14802	NaN	NaN	NaN	NaN	NaN	NaN	NaN
requires_job_training	25480	2	N	22525	NaN	NaN	NaN	NaN	NaN	NaN	NaN
no_of_employees	25480.0	NaN	NaN	NaN	5667.089207	22877.917453	11.0	1022.0	2109.0	3504.0	602069.0
yr_of_estab	25480.0	NaN	NaN	NaN	1979.409929	42.366929	1800.0	1976.0	1997.0	2005.0	2016.0
region_of_employment	25480	5	Northeast	7195	NaN	NaN	NaN	NaN	NaN	NaN	NaN
prevailing_wage	25480.0	NaN	NaN	NaN	74455.814592	52815.942327	2.1367	34015.48	70308.21	107735.5125	319210.27
unit_of_wage	25480	4	Year	22962	NaN	NaN	NaN	NaN	NaN	NaN	NaN
full_time_position	25480	2	Y	22773	NaN	NaN	NaN	NaN	NaN	NaN	NaN
case_status	25480	2	Certified	17018	NaN	NaN	NaN	NaN	NaN	NaN	NaN

Observation

1. Negative values are converted to positive values

Since, the prevailing_wages are according to the unit_of_wages. Converting them into one unit makes the comparision easier. Here, unit_of_wages column is not altered.

df1.loc[(df1['unit_of_wage'] == 'Hour'),"prevailing_wage"] = df1[df1['unit_of_wage'] == 'Hour']["prevailing_wage"] * 40*4*12
df1.loc[(df1['unit_of_wage'] == 'Week'),"prevailing_wage"] = df1[df1['unit_of_wage'] == 'Week']["prevailing_wage"] * 4*12
df1.loc[(df1['unit_of_wage'] == 'Month'),"prevailing_wage"] = df1[df1['unit_of_wage'] == 'Month']["prevailing_wage"] *12

df1[['prevailing_wage','unit_of_wage']]

	prevailing_wage	unit_of_wage
case_id
EZYV01	1137029.568	Hour
EZYV02	83425.650	Year
EZYV03	122996.860	Year
EZYV04	83434.030	Year
EZYV05	149907.390	Year
...	...	...
EZYV25476	77092.570	Year
EZYV25477	279174.790	Year
EZYV25478	146298.850	Year
EZYV25479	86154.770	Year
EZYV25480	70876.910	Year

25480 rows × 2 columns

df[['prevailing_wage','unit_of_wage']]

	prevailing_wage	unit_of_wage
case_id
EZYV01	592.2029	Hour
EZYV02	83425.6500	Year
EZYV03	122996.8600	Year
EZYV04	83434.0300	Year
EZYV05	149907.3900	Year
...	...	...
EZYV25476	77092.5700	Year
EZYV25477	279174.7900	Year
EZYV25478	146298.8500	Year
EZYV25479	86154.7700	Year
EZYV25480	70876.9100	Year

25480 rows × 2 columns

Observation

All the prevailing wages are converted to one unit. Keeping the unit_of_wages column same.

df1

	continent	education_of_employee	has_job_experience	requires_job_training	no_of_employees	yr_of_estab	region_of_employment	prevailing_wage	unit_of_wage	full_time_position	case_status
case_id
EZYV01	Asia	High School	N	N	14513	2007	West	1137029.568	Hour	Y	Denied
EZYV02	Asia	Master's	Y	N	2412	2002	Northeast	83425.650	Year	Y	Certified
EZYV03	Asia	Bachelor's	N	Y	44444	2008	West	122996.860	Year	Y	Denied
EZYV04	Asia	Bachelor's	N	N	98	1897	West	83434.030	Year	Y	Denied
EZYV05	Africa	Master's	Y	N	1082	2005	South	149907.390	Year	Y	Certified
...	...	...	...	...	...	...	...	...	...	...	...
EZYV25476	Asia	Bachelor's	Y	Y	2601	2008	South	77092.570	Year	Y	Certified
EZYV25477	Asia	High School	Y	N	3274	2006	Northeast	279174.790	Year	Y	Certified
EZYV25478	Asia	Master's	Y	N	1121	1910	South	146298.850	Year	N	Certified
EZYV25479	Asia	Master's	Y	Y	1918	1887	West	86154.770	Year	Y	Certified
EZYV25480	Asia	Bachelor's	Y	N	3195	1960	Midwest	70876.910	Year	Y	Certified

25480 rows × 11 columns

Outlier Treatment

# outlier detection using boxplot
numeric_columns = df1.select_dtypes(include=np.number).columns.tolist()

 #let's plot the boxplots of all columns to check for outliers
plt.figure(figsize=(20, 30))

for i, variable in enumerate(numeric_columns):
    plt.subplot(5, 4, i + 1)
    plt.boxplot(df1[variable], whis=1.5)
    plt.tight_layout()
    plt.title(variable)

plt.show()

Observation

1. There are many outliers among the no_of_employees, ye_of_estab and prevailing_wage. All need to be treated here.

# Outlier treatment
def treatment_outliers(df1, col):
    
    Q1 = df1[col].quantile(0.25)  # 25th quantile
    Q3 = df1[col].quantile(0.75)  # 75th quantile
    IQR = Q3 - Q1
    Lower_Whisker = Q1 - 1.5 * IQR
    Upper_Whisker = Q3 + 1.5 * IQR

    # all the values smaller than Lower_Whisker will be assigned the value of Lower_Whisker
    # all the values greater than Upper_Whisker will be assigned the value of Upper_Whisker
    df1[col] = np.clip(df1[col], Lower_Whisker, Upper_Whisker)

    return df1


def treat_all_outliers(df1, col_list):
    for c in col_list:
        df1 = treatment_outliers(df1, c)

    return df1

# treating the outliers
numerical_col = df1.select_dtypes(include=np.number).columns.tolist()
df1 = treat_all_outliers(df1, numerical_col)

# let's look at the boxplots to see if the outliers have been treated or not
plt.figure(figsize=(20, 30))

for i, variable in enumerate(numeric_columns):
    plt.subplot(5, 4, i + 1)
    plt.boxplot(df1[variable], whis=1.5)
    plt.tight_layout()
    plt.title(variable)

plt.show()

Observation

1. Outlier treatment is done.

# Lets look at the statistical summary of the data
df1.describe(include="all").T

	count	unique	top	freq	mean	std	min	25%	50%	75%	max
continent	25480	6	Asia	16861	NaN	NaN	NaN	NaN	NaN	NaN	NaN
education_of_employee	25480	4	Bachelor's	10234	NaN	NaN	NaN	NaN	NaN	NaN	NaN
has_job_experience	25480	2	Y	14802	NaN	NaN	NaN	NaN	NaN	NaN	NaN
requires_job_training	25480	2	N	22525	NaN	NaN	NaN	NaN	NaN	NaN	NaN
no_of_employees	25480.0	NaN	NaN	NaN	2495.034929	1874.429628	11.0	1022.0	2109.0	3504.0	7227.0
yr_of_estab	25480.0	NaN	NaN	NaN	1985.957143	25.813205	1932.5	1976.0	1997.0	2005.0	2016.0
region_of_employment	25480	5	Northeast	7195	NaN	NaN	NaN	NaN	NaN	NaN	NaN
prevailing_wage	25480.0	NaN	NaN	NaN	95533.802286	65328.842508	100.0	47092.79	82801.61	124783.27	241318.99
unit_of_wage	25480	4	Year	22962	NaN	NaN	NaN	NaN	NaN	NaN	NaN
full_time_position	25480	2	Y	22773	NaN	NaN	NaN	NaN	NaN	NaN	NaN
case_status	25480	2	Certified	17018	NaN	NaN	NaN	NaN	NaN	NaN	NaN

Data Preparation for Modeling

Converting the categorical variables to integers.

df1["case_status"].replace({"Certified": 1, "Denied": 0}, inplace=True)
df1["full_time_position"].replace({"Y": 1, "N": 0}, inplace=True)
df1["requires_job_training"].replace({"Y": 1, "N": 0}, inplace=True)
df1["has_job_experience"].replace({"Y": 1, "N": 0}, inplace=True)

# outlier detection using boxplot
numeric_columns = df1.select_dtypes(include=np.number).columns.tolist()

 #let's plot the boxplots of all columns to check for outliers
plt.figure(figsize=(20, 30))

for i, variable in enumerate(numeric_columns):
    plt.subplot(5, 4, i + 1)
    plt.boxplot(df1[variable], whis=1.5)
    plt.tight_layout()
    plt.title(variable)

plt.show()

df1

	continent	education_of_employee	has_job_experience	requires_job_training	no_of_employees	yr_of_estab	region_of_employment	prevailing_wage	unit_of_wage	full_time_position	case_status
case_id
EZYV01	Asia	High School	0	0	7227	2007.0	West	241318.99	Hour	1	0
EZYV02	Asia	Master's	1	0	2412	2002.0	Northeast	83425.65	Year	1	1
EZYV03	Asia	Bachelor's	0	1	7227	2008.0	West	122996.86	Year	1	0
EZYV04	Asia	Bachelor's	0	0	98	1932.5	West	83434.03	Year	1	0
EZYV05	Africa	Master's	1	0	1082	2005.0	South	149907.39	Year	1	1
...	...	...	...	...	...	...	...	...	...	...	...
EZYV25476	Asia	Bachelor's	1	1	2601	2008.0	South	77092.57	Year	1	1
EZYV25477	Asia	High School	1	0	3274	2006.0	Northeast	241318.99	Year	1	1
EZYV25478	Asia	Master's	1	0	1121	1932.5	South	146298.85	Year	0	1
EZYV25479	Asia	Master's	1	1	1918	1932.5	West	86154.77	Year	1	1
EZYV25480	Asia	Bachelor's	1	0	3195	1960.0	Midwest	70876.91	Year	1	1

25480 rows × 11 columns

df1.info()

<class 'pandas.core.frame.DataFrame'>
Index: 25480 entries, EZYV01 to EZYV25480
Data columns (total 11 columns):
 #   Column                 Non-Null Count  Dtype   
---  ------                 --------------  -----   
 0   continent              25480 non-null  category
 1   education_of_employee  25480 non-null  category
 2   has_job_experience     25480 non-null  int64   
 3   requires_job_training  25480 non-null  int64   
 4   no_of_employees        25480 non-null  int64   
 5   yr_of_estab            25480 non-null  float64 
 6   region_of_employment   25480 non-null  category
 7   prevailing_wage        25480 non-null  float64 
 8   unit_of_wage           25480 non-null  category
 9   full_time_position     25480 non-null  int64   
 10  case_status            25480 non-null  int64   
dtypes: category(4), float64(2), int64(5)
memory usage: 2.7+ MB

Observation

1. continent, education_of_employee, region_of_employment and unit_of_wage are categorical variables.
2. prevailing_wage is a float Vriable.
3. has_job_experience, requires_job_training and no_of_employees, full_time_position and case_status are integer variables.

EDA

• It is a good idea to explore the data once again after manipulating it.

Observations on no_of_employees

sns.histplot(df1.no_of_employees, kde = True)

<AxesSubplot:xlabel='no_of_employees', ylabel='Count'>

sns.boxplot(df1.no_of_employees,orient = "h");

Observation

1. The graph is positively skewed.
2. number of employees in most of the cases are below 2000.
3. Outliers are treated.

sns.histplot(df1.yr_of_estab, kde = True)

<AxesSubplot:xlabel='yr_of_estab', ylabel='Count'>

sns.boxplot(df1.yr_of_estab,orient = "h");

Observations.

1. Most of the companies are establieshed between 2000 and 2020.
2. The graph is negatively skewed.
3. The outliers are treated.

Observations on continent

labeled_barplot(df1, "continent", perc=True)

Observation

Most of the employees are from Asia. Very less number of employees are from Oceania.

Observations on education_of_employee

labeled_barplot(df1, "education_of_employee", perc=True)

Observation

40.2% of the employees have Bachelors degree. 37.8% of the employees have Master's degree. Very less number of employees(8.6%) of them have Doctor's degree.

Observation on has_job_experience

labeled_barplot(df1, "has_job_experience", perc=True)

Observation

58.1% of the applicants have job experience.

Observation on requires_job_training

labeled_barplot(df1, "requires_job_training", perc=True)

Observation

88.4% of the applicants donot require job training

Observations on region_of_employment

labeled_barplot(df1, "region_of_employment", perc=True)

Observation.

Very less(1.5%) of the applicants are from Island. Most of the employees are from Northeast, South and West regions.

Observations of unit_of_wage

labeled_barplot(df1, "unit_of_wage", perc=True)

Observation

90.1% of the employees receive year wages. 8.5% of the employees receive Hour wages. 0.3% of the employees receive Month wages. 1.1% of the employees receive Week wages.

Observations on full_time_position

labeled_barplot(df1, "full_time_position", perc=True)

Observation

89.4% of the employees are in full_time_position. Rest are not.

Observations on case_status

labeled_barplot(df1, "case_status", perc=True)

Observation

66.8% of the cases are Certified.

Bivariate Analysis

Plotting the Bivariate Analysis to understand the interaction with each other

sns.heatmap(df1.corr(),cmap="YlGnBu",vmin=-1,vmax=1, annot=True)

<AxesSubplot:>

Observation

1. Highest correlation from the above heatmap is case_status and has_job_experience followed by requires_job_training and full_time_position.
2. It is important to note that correlation does not imply causation.
3. Some are negatively correlated.

Bivariate Scatter Plots

sns.pairplot(df1);

Observation

Highest correlation from the above heatmap is case_status and has_job_experience followed by requires_job_training and full_time_position. It is important to note that correlation does not imply causation. Some are negatively correlated.

Check continent by case_status

sns.countplot(x="continent", hue="case_status", data=df1, palette='Set1',saturation=50 );

Observation

Most of the applicants whose case status is certified are from Asia. Most of the applicants whose case status is Denied are also from Asia. Least rejections are for those aplicants from Oceania.

Check education_of_employee count by case_status

sns.countplot(x="education_of_employee", hue="case_status", data=df1, palette='Set1',saturation=50 );

Observation

Most of the applicants whose case status is certified have Master's forllowed by candidates with Bachelor's degree. Rejections are very less among Doctorate candidates.

Check has_job_experience by case_count

sns.countplot(x="has_job_experience", hue="case_status", data=df1, palette='Set1',saturation=50 );

Observations

Most of the applicants whose case_status is certified have job experience. There are more rejections among applicants whose case_status is denied.

check requires_job_training by case_status

sns.countplot(x="requires_job_training", hue="case_status", data=df1, palette='Set1',saturation=50 );

Observations

Most of the candidates whose case is certified are those who did not require job training. Very less rejections are among those who required job training.

check region_of_employment by case_status

sns.countplot(x="region_of_employment", hue="case_status", data=df1, palette='Set1',saturation=50 );

Observation

Most of the certified applicantions are from south region followed by Northeast and West. Very less rejected applications are from Island.

check unit_of_wage by case_count

sns.countplot(x="unit_of_wage", hue="case_status", data=df1, palette='Set1',saturation=50 );

Observation

Most of the employees receive yearly wages. The maximum cases are Certified among Yearly waged applicants. Least rejections are among Monthly waged and weekly waged applicants.

Check full_time_position by case_status

sns.countplot(x="full_time_position", hue="case_status", data=df1, palette='Set1',saturation=50 );

Observation

Most of the certified applicants have full_time_position. Very less rejected candidates have full_time_position.

Relationship between no_of_employees and yr_of_estab

fig = plt.figure(figsize= (10,5))
ax = sns.regplot(x ='no_of_employees', y = df1['yr_of_estab'], data = df1)

Observation

Companies established between 2000 and 2020 have higher employee count.

Relationship between yr_of_estab and prevailing_wage

fig = plt.figure(figsize= (10,5))
ax = sns.regplot(x ='yr_of_estab', y = df1['prevailing_wage'], data = df1)

Observation

High Prevailing wages are among companies established between 1975 and 2000.

Relationship between no_of_employees and prevailing_wage

fig = plt.figure(figsize= (10,5))
ax = sns.scatterplot(x ='no_of_employees', y = df1['prevailing_wage'], data = df1)

Observation

1. More prevailing wages are given to the employee whose company employee count is less than 5000.

Correlation between no_of_employees, yr_of_estab and case_status

figure = plt.figure(figsize=(8,7))
sns.scatterplot(x="no_of_employees", y="yr_of_estab", data=df1, hue='case_status', palette='tab10' )
plt.show()

Observation

1. Case_status = 1 when company's year of establishment is between 1980 and 2020 and employee count less than 5000.

Correlation between yr_of_estab, prevailing_wage and case_status

figure = plt.figure(figsize=(8,7))
sns.scatterplot(x="prevailing_wage", y="yr_of_estab", data=df1, hue='case_status', palette='tab10' )
plt.show()

Observation

1. Accpetance criteria is more when prevailing_wage is less than 15000 and company year of establishment is between 1980 and 2020.

Correlation between no_of_employees, prevailing_wage and case_status

figure = plt.figure(figsize=(8,7))
sns.scatterplot(x="prevailing_wage", y="no_of_employees", data=df1, hue='case_status', palette='tab10' )
plt.show()

Observation

1. Visa status acceptance criteria is more when the company's employee count is less than 5000 and prevailing_wage is less than 150000.

Correlation between no_of_employees, continent and case_status

fig = plt.figure(figsize= (15,5))
ax = sns.barplot(x ='continent', y = 'no_of_employees', data = df1, hue = 'case_status')

Observation

After outlier treatment, there is a change in the graph, The employee belonging to the companies with higher employee count and region - Africa has highest acceptance. The employee belonging to the companies with less employee count and region - Oceania has highest rejections.

1. Those with higher education may want to travel abroad for a well-paid job. Does education play a role in Visa certification?

labeled_barplot(df1, "education_of_employee", perc=True)

sns.countplot(x="education_of_employee", hue="case_status", data=df1, palette='Set1',saturation=50 );

Observation

In the given dataset, there are more applicants with Bachelor's degree followed by Master's degree. More cases are certified among the employees who have Master's degree followed by Bachelor's degree. Least rejections are among the employees who have Doctorate degree.

2. How does the visa status vary across different continents?

sns.countplot(x="continent", hue="case_status", data=df1, palette='Set1',saturation=50 );

Observation

More cases are certified among the employees who belong to Asia. Least cases are certified among the employees who belong to Oceania. Less rejections are made among the employees who belong to Africa followed by South America.

fig = plt.figure(figsize= (15,5))
ax = sns.barplot(x ='continent', y = 'no_of_employees', data = df1, hue = 'case_status')

Observation

1. After outlier treatment, there is a change in the graph,
2. The employee belonging to the companies with higher employee count and region - Africa has highest acceptance.
3. The employee belonging to the companies with less employee count and region - Oceania has highest rejections.

3. Experienced professionals might look abroad for opportunities to improve their lifestyles and career development. Does work experience influence visa status

labeled_barplot(df1, "has_job_experience", perc=True)

sns.countplot(x="has_job_experience", hue="case_status", data=df1, palette='Set1',saturation=50 );

Observation

From the given dataset, there are more employees who have job experience. More cases are certified among the employees who have job exoerience. More rejections are among the employees who donot have job experience.

fig = plt.figure(figsize= (15,5))
ax = sns.barplot(x ='has_job_experience', y = 'no_of_employees', data = df1, hue = 'case_status')

Observation

Aplicants from companies with more employee count and has job experience have higher chances of being accepted.

4. In the United States, employees are paid at different intervals. Which pay unit is most likely to be certified for a visa?

sns.countplot(x="unit_of_wage", hue="case_status", data=df1, palette='Set1',saturation=50 );

Observation

Most of the employees receive yearly wages. The maximum cases are Certified among Yearly waged applicants. Least rejections are among Monthly waged and weekly waged applicants.

5. The US government has established a prevailing wage to protect local talent and foreign workers. How does the visa status change with the prevailing wage?

plt.figure(figsize=(15,5))

ax = sns.barplot(x='case_status', y='prevailing_wage', data=df1)

Observation

Certified applicants have more prevailing_wages than rejected applicants.

Checking Multicollinearity

X = df1.drop(['case_status'], axis = 1) 
y = df1['case_status']

X.head()

	continent	education_of_employee	has_job_experience	requires_job_training	no_of_employees	yr_of_estab	region_of_employment	prevailing_wage	unit_of_wage	full_time_position
case_id
EZYV01	Asia	High School	0	0	7227	2007.0	West	241318.99	Hour	1
EZYV02	Asia	Master's	1	0	2412	2002.0	Northeast	83425.65	Year	1
EZYV03	Asia	Bachelor's	0	1	7227	2008.0	West	122996.86	Year	1
EZYV04	Asia	Bachelor's	0	0	98	1932.5	West	83434.03	Year	1
EZYV05	Africa	Master's	1	0	1082	2005.0	South	149907.39	Year	1

y.head()

case_id
EZYV01    0
EZYV02    1
EZYV03    0
EZYV04    0
EZYV05    1
Name: case_status, dtype: int64

X = pd.get_dummies(X, columns = X.select_dtypes(include = ["object","category"]).columns.tolist(), drop_first = True)

X.head()

	has_job_experience	requires_job_training	no_of_employees	yr_of_estab	prevailing_wage	full_time_position	continent_Asia	continent_Europe	continent_North America	continent_Oceania	continent_South America	education_of_employee_Doctorate	education_of_employee_High School	education_of_employee_Master's	region_of_employment_Midwest	region_of_employment_Northeast	region_of_employment_South	region_of_employment_West	unit_of_wage_Month	unit_of_wage_Week	unit_of_wage_Year
case_id
EZYV01	0	0	7227	2007.0	241318.99	1	1	0	0	0	0	0	1	0	0	0	0	1	0	0	0
EZYV02	1	0	2412	2002.0	83425.65	1	1	0	0	0	0	0	0	1	0	1	0	0	0	0	1
EZYV03	0	1	7227	2008.0	122996.86	1	1	0	0	0	0	0	0	0	0	0	0	1	0	0	1
EZYV04	0	0	98	1932.5	83434.03	1	1	0	0	0	0	0	0	0	0	0	0	1	0	0	1
EZYV05	1	0	1082	2005.0	149907.39	1	0	0	0	0	0	0	0	1	0	0	1	0	0	0	1

y.head()

case_id
EZYV01    0
EZYV02    1
EZYV03    0
EZYV04    0
EZYV05    1
Name: case_status, dtype: int64

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=.30, random_state=1,stratify=y)

from statsmodels.stats.outliers_influence import variance_inflation_factor

def checking_vif(predictors):
    vif = pd.DataFrame()
    vif["feature"] = predictors.columns
    vif["VIF"] = [variance_inflation_factor(predictors.values,i)
                 for i in range(len(predictors.columns))
    ]
    return vif

Building the model

Model Evaluation Criterion

Model can make wrong predictions as:

1. Predicting a case_status gets certified but in reality the OFLC rejects the case.
2. Predicting a case_status gets rejected, the case will never go to OFLC.

Which case is more important?

Both the cases are important.

1. Predicting a case_status doesnot get certified but in reality OFLC rejects as this might result in huge loss leading to number of negative applications, huge time loss and resource crunch.
2. Predicting a case_status gets rejected, the case will never go to OFLC. This will result in loosing a good candidate. Since there are many applications, loosing some might not affect much.

How to reduce this loss i.e need to reduce False Negatives?

1. Recall gives the ratio of True posiitives to Actual Positives. So, high recall means less false negatives. Lower chances of predicting the certified as rejected.
2. Recall should be maximized, the greater the Recall higher the chances minimizing false negatives. ##### First, let's create functions to calculate different metrics and confusion matrix so that we don't have to use the same code repeatedly for each model.
3. The make_confusion_matrix_for_model function will be used to plot confusion matrix.
4. get_metrics_score fuction is used to calculate the metrics.

## Function to create confusion matrix
def make_confusion_matrix_for_model(model,y_actual,labels=[1, 0]):
    y_predict = model.predict(X_test)
    cm=metrics.confusion_matrix( y_actual, y_predict, labels=[0, 1])
    df_cm = pd.DataFrame(cm, index = [i for i in ["Actual - No","Actual - Yes"]],
                  columns = [i for i in ['Predicted - No','Predicted - Yes']])
    group_counts = ["{0:0.0f}".format(value) for value in
                cm.flatten()]
    group_percentages = ["{0:.2%}".format(value) for value in
                         cm.flatten()/np.sum(cm)]
    labels = [f"{v1}\n{v2}" for v1, v2 in
              zip(group_counts,group_percentages)]
    labels = np.asarray(labels).reshape(2,2)
    plt.figure(figsize = (10,7))
    sns.heatmap(df_cm, annot=labels,fmt='')
    plt.ylabel('True label')
    plt.xlabel('Predicted label')

##  Function to calculate different metric scores of the model - Accuracy, Recall and Precision
def get_metrics_score(model,flag=True):
    '''
    model : classifier to predict values of X

    '''
    # defining an empty list to store train and test results
    score_list=[] 
    
    #Predicting on train and tests
    pred_train = model.predict(X_train)
    pred_test = model.predict(X_test)
    
    #Accuracy of the model
    train_acc = model.score(X_train,y_train)
    test_acc = model.score(X_test,y_test)
    
    #Recall of the model
    train_recall = metrics.recall_score(y_train,pred_train)
    test_recall = metrics.recall_score(y_test,pred_test)
    
    #Precision of the model
    train_precision = metrics.precision_score(y_train,pred_train)
    test_precision = metrics.precision_score(y_test,pred_test)
    
    #f-score of the model
    train_fscore = metrics.f1_score(y_train,pred_train)
    test_fscore = metrics.f1_score(y_test,pred_test)
    
    score_list.extend((train_acc,test_acc,train_recall,test_recall,train_precision,test_precision,train_fscore,test_fscore))
        
    # If the flag is set to True then only the following print statements will be dispayed. The default value is set to True.
    if flag == True: 
        print("Accuracy on training set : ",model.score(X_train,y_train))
        print("Accuracy on test set : ",model.score(X_test,y_test))
        print("Recall on training set : ",metrics.recall_score(y_train,pred_train))
        print("Recall on test set : ",metrics.recall_score(y_test,pred_test))
        print("Precision on training set : ",metrics.precision_score(y_train,pred_train))
        print("Precision on test set : ",metrics.precision_score(y_test,pred_test))
        print("f1-score on training set : ",metrics.f1_score(y_train,pred_train))
        print("f1-score on test set : ",metrics.f1_score(y_test,pred_test))
 
    return score_list # returning the list with train and test scores

Building a Decision Tree model

1. We will build our model using the DecisionTreeClassifier function.
2. Using default 'gini' criteria to split. Other option include 'entropy'.

dTree = DecisionTreeClassifier(criterion = 'gini', random_state=1)
dTree.fit(X_train, y_train)

DecisionTreeClassifier(random_state=1)

Confusion matrix

make_confusion_matrix_for_model(dTree,y_test)

#Using above defined function to get accuracy, recall and precision on train and test set
dTree_score=get_metrics_score(dTree)

Accuracy on training set :  1.0
Accuracy on test set :  0.6593406593406593
Recall on training set :  1.0
Recall on test set :  0.7380999020568071
Precision on training set :  1.0
Precision on test set :  0.7483614697120159
f1-score on training set :  1.0
f1-score on test set :  0.7431952662721892

Visualize the dTree

feature_names = list(X.columns)
print(feature_names)

['has_job_experience', 'requires_job_training', 'no_of_employees', 'yr_of_estab', 'prevailing_wage', 'full_time_position', 'continent_Asia', 'continent_Europe', 'continent_North America', 'continent_Oceania', 'continent_South America', 'education_of_employee_Doctorate', 'education_of_employee_High School', "education_of_employee_Master's", 'region_of_employment_Midwest', 'region_of_employment_Northeast', 'region_of_employment_South', 'region_of_employment_West', 'unit_of_wage_Month', 'unit_of_wage_Week', 'unit_of_wage_Year']

# Text report showing the rules of a decision tree -

print(tree.export_text(dTree,feature_names=feature_names,show_weights=True))

|--- education_of_employee_High School <= 0.50
|   |--- has_job_experience <= 0.50
|   |   |--- unit_of_wage_Year <= 0.50
|   |   |   |--- education_of_employee_Doctorate <= 0.50
|   |   |   |   |--- education_of_employee_Master's <= 0.50
|   |   |   |   |   |--- unit_of_wage_Week <= 0.50
|   |   |   |   |   |   |--- requires_job_training <= 0.50
|   |   |   |   |   |   |   |--- unit_of_wage_Month <= 0.50
|   |   |   |   |   |   |   |   |--- no_of_employees <= 256.00
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 26.00
|   |   |   |   |   |   |   |   |   |   |--- continent_North America <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- continent_North America >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  26.00
|   |   |   |   |   |   |   |   |   |   |--- region_of_employment_Northeast <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [14.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- region_of_employment_Northeast >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |--- no_of_employees >  256.00
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 349.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 321.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  321.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 3.00] class: 1
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  349.50
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 2003.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 21
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab >  2003.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 12
|   |   |   |   |   |   |   |--- unit_of_wage_Month >  0.50
|   |   |   |   |   |   |   |   |--- no_of_employees <= 1500.50
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |   |   |   |--- no_of_employees >  1500.50
|   |   |   |   |   |   |   |   |   |--- region_of_employment_South <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 2008.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab >  2008.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- region_of_employment_South >  0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |--- requires_job_training >  0.50
|   |   |   |   |   |   |   |--- no_of_employees <= 2458.50
|   |   |   |   |   |   |   |   |--- unit_of_wage_Month <= 0.50
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 6.00] class: 1
|   |   |   |   |   |   |   |   |--- unit_of_wage_Month >  0.50
|   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- no_of_employees >  2458.50
|   |   |   |   |   |   |   |   |--- weights: [5.00, 0.00] class: 0
|   |   |   |   |   |--- unit_of_wage_Week >  0.50
|   |   |   |   |   |   |--- region_of_employment_South <= 0.50
|   |   |   |   |   |   |   |--- prevailing_wage <= 173057.01
|   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |--- prevailing_wage >  173057.01
|   |   |   |   |   |   |   |   |--- requires_job_training <= 0.50
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 5544.50
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 2013.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab >  2013.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  5544.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |--- requires_job_training >  0.50
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |--- region_of_employment_South >  0.50
|   |   |   |   |   |   |   |--- no_of_employees <= 2983.00
|   |   |   |   |   |   |   |   |--- weights: [0.00, 4.00] class: 1
|   |   |   |   |   |   |   |--- no_of_employees >  2983.00
|   |   |   |   |   |   |   |   |--- no_of_employees <= 3407.00
|   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- no_of_employees >  3407.00
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |--- education_of_employee_Master's >  0.50
|   |   |   |   |   |--- continent_North America <= 0.50
|   |   |   |   |   |   |--- unit_of_wage_Week <= 0.50
|   |   |   |   |   |   |   |--- yr_of_estab <= 1996.50
|   |   |   |   |   |   |   |   |--- prevailing_wage <= 206545.16
|   |   |   |   |   |   |   |   |   |--- weights: [8.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- prevailing_wage >  206545.16
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 4647.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 3892.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 13
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  3892.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  4647.50
|   |   |   |   |   |   |   |   |   |   |--- continent_Oceania <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- continent_Oceania >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |--- yr_of_estab >  1996.50
|   |   |   |   |   |   |   |   |--- yr_of_estab <= 1997.50
|   |   |   |   |   |   |   |   |   |--- weights: [8.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- yr_of_estab >  1997.50
|   |   |   |   |   |   |   |   |   |--- region_of_employment_South <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 2007.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 11
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab >  2007.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 6
|   |   |   |   |   |   |   |   |   |--- region_of_employment_South >  0.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 771.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  771.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 7
|   |   |   |   |   |   |--- unit_of_wage_Week >  0.50
|   |   |   |   |   |   |   |--- no_of_employees <= 664.50
|   |   |   |   |   |   |   |   |--- no_of_employees <= 416.00
|   |   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- no_of_employees >  416.00
|   |   |   |   |   |   |   |   |   |--- region_of_employment_West <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |--- region_of_employment_West >  0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- no_of_employees >  664.50
|   |   |   |   |   |   |   |   |--- weights: [0.00, 9.00] class: 1
|   |   |   |   |   |--- continent_North America >  0.50
|   |   |   |   |   |   |--- prevailing_wage <= 10229.09
|   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |--- prevailing_wage >  10229.09
|   |   |   |   |   |   |   |--- region_of_employment_West <= 0.50
|   |   |   |   |   |   |   |   |--- yr_of_estab <= 2010.50
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 823.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [9.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  823.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 916.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  916.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 8
|   |   |   |   |   |   |   |   |--- yr_of_estab >  2010.50
|   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 2012.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 1496.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  1496.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |--- yr_of_estab >  2012.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- region_of_employment_West >  0.50
|   |   |   |   |   |   |   |   |--- no_of_employees <= 1936.00
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 795.00
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 387.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  387.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  795.00
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |   |   |   |--- no_of_employees >  1936.00
|   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 2006.00
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 3274.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  3274.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [5.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- yr_of_estab >  2006.00
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |--- education_of_employee_Doctorate >  0.50
|   |   |   |   |--- region_of_employment_West <= 0.50
|   |   |   |   |   |--- yr_of_estab <= 1955.00
|   |   |   |   |   |   |--- no_of_employees <= 2925.00
|   |   |   |   |   |   |   |--- continent_Europe <= 0.50
|   |   |   |   |   |   |   |   |--- weights: [4.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- continent_Europe >  0.50
|   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |--- no_of_employees >  2925.00
|   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |--- yr_of_estab >  1955.00
|   |   |   |   |   |   |--- continent_Asia <= 0.50
|   |   |   |   |   |   |   |--- yr_of_estab <= 1984.50
|   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- yr_of_estab >  1984.50
|   |   |   |   |   |   |   |   |--- no_of_employees <= 653.50
|   |   |   |   |   |   |   |   |   |--- continent_North America <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 317.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  317.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- continent_North America >  0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- no_of_employees >  653.50
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 6.00] class: 1
|   |   |   |   |   |   |--- continent_Asia >  0.50
|   |   |   |   |   |   |   |--- yr_of_estab <= 1994.50
|   |   |   |   |   |   |   |   |--- weights: [0.00, 10.00] class: 1
|   |   |   |   |   |   |   |--- yr_of_estab >  1994.50
|   |   |   |   |   |   |   |   |--- yr_of_estab <= 1996.00
|   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- yr_of_estab >  1996.00
|   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 2000.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 999.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  999.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |--- yr_of_estab >  2000.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 3.00] class: 1
|   |   |   |   |--- region_of_employment_West >  0.50
|   |   |   |   |   |--- yr_of_estab <= 1970.00
|   |   |   |   |   |   |--- weights: [0.00, 3.00] class: 1
|   |   |   |   |   |--- yr_of_estab >  1970.00
|   |   |   |   |   |   |--- no_of_employees <= 1633.00
|   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |--- no_of_employees >  1633.00
|   |   |   |   |   |   |   |--- unit_of_wage_Week <= 0.50
|   |   |   |   |   |   |   |   |--- weights: [8.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- unit_of_wage_Week >  0.50
|   |   |   |   |   |   |   |   |--- no_of_employees <= 4036.50
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |--- no_of_employees >  4036.50
|   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |--- unit_of_wage_Year >  0.50
|   |   |   |--- continent_Europe <= 0.50
|   |   |   |   |--- region_of_employment_Midwest <= 0.50
|   |   |   |   |   |--- region_of_employment_South <= 0.50
|   |   |   |   |   |   |--- education_of_employee_Doctorate <= 0.50
|   |   |   |   |   |   |   |--- education_of_employee_Master's <= 0.50
|   |   |   |   |   |   |   |   |--- full_time_position <= 0.50
|   |   |   |   |   |   |   |   |   |--- weights: [95.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- full_time_position >  0.50
|   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 2004.50
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 217638.20
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 26
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  217638.20
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |   |--- yr_of_estab >  2004.50
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 147922.91
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 19
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  147922.91
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |--- education_of_employee_Master's >  0.50
|   |   |   |   |   |   |   |   |--- full_time_position <= 0.50
|   |   |   |   |   |   |   |   |   |--- requires_job_training <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 231944.81
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 15
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  231944.81
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 11.00] class: 1
|   |   |   |   |   |   |   |   |   |--- requires_job_training >  0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [4.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- full_time_position >  0.50
|   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 56476.72
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 368.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  368.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 15
|   |   |   |   |   |   |   |   |   |--- prevailing_wage >  56476.72
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 64817.39
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 9
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  64817.39
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 18
|   |   |   |   |   |   |--- education_of_employee_Doctorate >  0.50
|   |   |   |   |   |   |   |--- no_of_employees <= 84.00
|   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- no_of_employees >  84.00
|   |   |   |   |   |   |   |   |--- prevailing_wage <= 61070.79
|   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 58162.18
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 439.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  439.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 8
|   |   |   |   |   |   |   |   |   |--- prevailing_wage >  58162.18
|   |   |   |   |   |   |   |   |   |   |--- weights: [4.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- prevailing_wage >  61070.79
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 1011.00
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 889.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 6
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  889.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  1011.00
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 3054.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  3054.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 7
|   |   |   |   |   |--- region_of_employment_South >  0.50
|   |   |   |   |   |   |--- requires_job_training <= 0.50
|   |   |   |   |   |   |   |--- full_time_position <= 0.50
|   |   |   |   |   |   |   |   |--- no_of_employees <= 1687.50
|   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 2011.00
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 1932.75
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab >  1932.75
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 10
|   |   |   |   |   |   |   |   |   |--- yr_of_estab >  2011.00
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 580.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  580.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |--- no_of_employees >  1687.50
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 3372.00
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 168822.02
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 12
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  168822.02
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 12.00] class: 1
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  3372.00
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 6703.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 8
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  6703.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |--- full_time_position >  0.50
|   |   |   |   |   |   |   |   |--- no_of_employees <= 3438.00
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 3341.00
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 129022.18
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 22
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  129022.18
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 10
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  3341.00
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 26857.03
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  26857.03
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |--- no_of_employees >  3438.00
|   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 2002.50
|   |   |   |   |   |   |   |   |   |   |--- continent_North America <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 12
|   |   |   |   |   |   |   |   |   |   |--- continent_North America >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |   |--- yr_of_estab >  2002.50
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 2006.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab >  2006.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 7
|   |   |   |   |   |   |--- requires_job_training >  0.50
|   |   |   |   |   |   |   |--- full_time_position <= 0.50
|   |   |   |   |   |   |   |   |--- weights: [4.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- full_time_position >  0.50
|   |   |   |   |   |   |   |   |--- education_of_employee_Doctorate <= 0.50
|   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 41991.57
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 1996.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 6
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab >  1996.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 6
|   |   |   |   |   |   |   |   |   |--- prevailing_wage >  41991.57
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 25.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  25.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 19
|   |   |   |   |   |   |   |   |--- education_of_employee_Doctorate >  0.50
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 93.50
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 96273.66
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  96273.66
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  93.50
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 109519.62
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 29.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  109519.62
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |--- region_of_employment_Midwest >  0.50
|   |   |   |   |   |--- yr_of_estab <= 2004.50
|   |   |   |   |   |   |--- yr_of_estab <= 1983.50
|   |   |   |   |   |   |   |--- prevailing_wage <= 146230.30
|   |   |   |   |   |   |   |   |--- no_of_employees <= 42.50
|   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- no_of_employees >  42.50
|   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 146073.00
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 7668.12
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  7668.12
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 17
|   |   |   |   |   |   |   |   |   |--- prevailing_wage >  146073.00
|   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- prevailing_wage >  146230.30
|   |   |   |   |   |   |   |   |--- continent_South America <= 0.50
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 6771.00
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 1979.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 41.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab >  1979.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  6771.00
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 161508.94
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 3.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  161508.94
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |--- continent_South America >  0.50
|   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 205387.09
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |--- prevailing_wage >  205387.09
|   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |--- yr_of_estab >  1983.50
|   |   |   |   |   |   |   |--- education_of_employee_Master's <= 0.50
|   |   |   |   |   |   |   |   |--- prevailing_wage <= 52797.89
|   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 49531.99
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 12634.42
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 4.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  12634.42
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 8
|   |   |   |   |   |   |   |   |   |--- prevailing_wage >  49531.99
|   |   |   |   |   |   |   |   |   |   |--- weights: [5.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- prevailing_wage >  52797.89
|   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 84548.84
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 3941.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 6
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  3941.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |--- prevailing_wage >  84548.84
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 1158.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  1158.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 8
|   |   |   |   |   |   |   |--- education_of_employee_Master's >  0.50
|   |   |   |   |   |   |   |   |--- yr_of_estab <= 1990.50
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 661.00
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 203.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  203.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [4.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  661.00
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 1618.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  1618.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 7
|   |   |   |   |   |   |   |   |--- yr_of_estab >  1990.50
|   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 50985.89
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 429.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  429.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |   |--- prevailing_wage >  50985.89
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 52847.39
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [3.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  52847.39
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 23
|   |   |   |   |   |--- yr_of_estab >  2004.50
|   |   |   |   |   |   |--- prevailing_wage <= 9198.22
|   |   |   |   |   |   |   |--- no_of_employees <= 1653.00
|   |   |   |   |   |   |   |   |--- weights: [3.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- no_of_employees >  1653.00
|   |   |   |   |   |   |   |   |--- no_of_employees <= 4925.50
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 3.00] class: 1
|   |   |   |   |   |   |   |   |--- no_of_employees >  4925.50
|   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |--- prevailing_wage >  9198.22
|   |   |   |   |   |   |   |--- yr_of_estab <= 2014.50
|   |   |   |   |   |   |   |   |--- continent_Oceania <= 0.50
|   |   |   |   |   |   |   |   |   |--- requires_job_training <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 784.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  784.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 15
|   |   |   |   |   |   |   |   |   |--- requires_job_training >  0.50
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 2010.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab >  2010.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [3.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- continent_Oceania >  0.50
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 3284.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  3284.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |--- yr_of_estab >  2014.50
|   |   |   |   |   |   |   |   |--- no_of_employees <= 1386.50
|   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 142347.97
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |   |   |   |   |--- prevailing_wage >  142347.97
|   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- no_of_employees >  1386.50
|   |   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |--- continent_Europe >  0.50
|   |   |   |   |--- prevailing_wage <= 27125.25
|   |   |   |   |   |--- no_of_employees <= 1347.50
|   |   |   |   |   |   |--- weights: [0.00, 49.00] class: 1
|   |   |   |   |   |--- no_of_employees >  1347.50
|   |   |   |   |   |   |--- no_of_employees <= 1402.50
|   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |--- no_of_employees >  1402.50
|   |   |   |   |   |   |   |--- no_of_employees <= 2039.00
|   |   |   |   |   |   |   |   |--- no_of_employees <= 2037.00
|   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 2006.50
|   |   |   |   |   |   |   |   |   |   |--- region_of_employment_Midwest <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 19.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- region_of_employment_Midwest >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |--- yr_of_estab >  2006.50
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 9680.11
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 3.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  9680.11
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |--- no_of_employees >  2037.00
|   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- no_of_employees >  2039.00
|   |   |   |   |   |   |   |   |--- no_of_employees <= 4284.50
|   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 1593.03
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 3023.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  3023.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- prevailing_wage >  1593.03
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 62.00] class: 1
|   |   |   |   |   |   |   |   |--- no_of_employees >  4284.50
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 4357.00
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 25782.11
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  25782.11
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  4357.00
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 1995.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 7
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab >  1995.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 29.00] class: 1
|   |   |   |   |--- prevailing_wage >  27125.25
|   |   |   |   |   |--- prevailing_wage <= 27214.77
|   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |--- prevailing_wage >  27214.77
|   |   |   |   |   |   |--- no_of_employees <= 52.00
|   |   |   |   |   |   |   |--- yr_of_estab <= 2001.50
|   |   |   |   |   |   |   |   |--- prevailing_wage <= 53593.63
|   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 34283.16
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |--- prevailing_wage >  34283.16
|   |   |   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- prevailing_wage >  53593.63
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 3.00] class: 1
|   |   |   |   |   |   |   |--- yr_of_estab >  2001.50
|   |   |   |   |   |   |   |   |--- weights: [3.00, 0.00] class: 0
|   |   |   |   |   |   |--- no_of_employees >  52.00
|   |   |   |   |   |   |   |--- education_of_employee_Doctorate <= 0.50
|   |   |   |   |   |   |   |   |--- prevailing_wage <= 127061.20
|   |   |   |   |   |   |   |   |   |--- region_of_employment_Northeast <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- region_of_employment_Midwest <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 10
|   |   |   |   |   |   |   |   |   |   |--- region_of_employment_Midwest >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 7
|   |   |   |   |   |   |   |   |   |--- region_of_employment_Northeast >  0.50
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 1945.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 10.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab >  1945.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 17
|   |   |   |   |   |   |   |   |--- prevailing_wage >  127061.20
|   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 131520.57
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 3733.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  3733.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |--- prevailing_wage >  131520.57
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 1979.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab >  1979.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 9
|   |   |   |   |   |   |   |--- education_of_employee_Doctorate >  0.50
|   |   |   |   |   |   |   |   |--- no_of_employees <= 5583.50
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 1285.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 1259.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 9
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  1259.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  1285.50
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 29598.92
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  29598.92
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 7
|   |   |   |   |   |   |   |   |--- no_of_employees >  5583.50
|   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 46579.42
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 5.00] class: 1
|   |   |   |   |   |   |   |   |   |--- prevailing_wage >  46579.42
|   |   |   |   |   |   |   |   |   |   |--- region_of_employment_Northeast <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |   |--- region_of_employment_Northeast >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [3.00, 0.00] class: 0
|   |--- has_job_experience >  0.50
|   |   |--- education_of_employee_Master's <= 0.50
|   |   |   |--- education_of_employee_Doctorate <= 0.50
|   |   |   |   |--- unit_of_wage_Year <= 0.50
|   |   |   |   |   |--- unit_of_wage_Week <= 0.50
|   |   |   |   |   |   |--- unit_of_wage_Month <= 0.50
|   |   |   |   |   |   |   |--- no_of_employees <= 1469.50
|   |   |   |   |   |   |   |   |--- prevailing_wage <= 223248.00
|   |   |   |   |   |   |   |   |   |--- weights: [14.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- prevailing_wage >  223248.00
|   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 233616.00
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |--- prevailing_wage >  233616.00
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 2011.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 11
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab >  2011.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |--- no_of_employees >  1469.50
|   |   |   |   |   |   |   |   |--- no_of_employees <= 2724.50
|   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 2001.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 2528.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 13
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  2528.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 5.00] class: 1
|   |   |   |   |   |   |   |   |   |--- yr_of_estab >  2001.50
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 2007.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab >  2007.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |--- no_of_employees >  2724.50
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 2916.00
|   |   |   |   |   |   |   |   |   |   |--- weights: [10.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  2916.00
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 1946.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab >  1946.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 10
|   |   |   |   |   |   |--- unit_of_wage_Month >  0.50
|   |   |   |   |   |   |   |--- yr_of_estab <= 1996.50
|   |   |   |   |   |   |   |   |--- region_of_employment_West <= 0.50
|   |   |   |   |   |   |   |   |   |--- region_of_employment_South <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 9.00] class: 1
|   |   |   |   |   |   |   |   |   |--- region_of_employment_South >  0.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 643.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  643.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- region_of_employment_West >  0.50
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 427.00
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  427.00
|   |   |   |   |   |   |   |   |   |   |--- continent_Asia <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- continent_Asia >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- yr_of_estab >  1996.50
|   |   |   |   |   |   |   |   |--- yr_of_estab <= 1999.50
|   |   |   |   |   |   |   |   |   |--- weights: [3.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- yr_of_estab >  1999.50
|   |   |   |   |   |   |   |   |   |--- region_of_employment_Midwest <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 3686.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 4.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  3686.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- region_of_employment_Midwest >  0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |--- unit_of_wage_Week >  0.50
|   |   |   |   |   |   |--- yr_of_estab <= 1996.50
|   |   |   |   |   |   |   |--- yr_of_estab <= 1994.00
|   |   |   |   |   |   |   |   |--- requires_job_training <= 0.50
|   |   |   |   |   |   |   |   |   |--- region_of_employment_West <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 2629.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  2629.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |   |--- region_of_employment_West >  0.50
|   |   |   |   |   |   |   |   |   |   |--- full_time_position <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- full_time_position >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |--- requires_job_training >  0.50
|   |   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- yr_of_estab >  1994.00
|   |   |   |   |   |   |   |   |--- weights: [4.00, 0.00] class: 0
|   |   |   |   |   |   |--- yr_of_estab >  1996.50
|   |   |   |   |   |   |   |--- full_time_position <= 0.50
|   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- full_time_position >  0.50
|   |   |   |   |   |   |   |   |--- continent_North America <= 0.50
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 12.00] class: 1
|   |   |   |   |   |   |   |   |--- continent_North America >  0.50
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 3051.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 1267.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  1267.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [3.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  3051.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 4.00] class: 1
|   |   |   |   |--- unit_of_wage_Year >  0.50
|   |   |   |   |   |--- continent_Europe <= 0.50
|   |   |   |   |   |   |--- continent_Asia <= 0.50
|   |   |   |   |   |   |   |--- prevailing_wage <= 7522.50
|   |   |   |   |   |   |   |   |--- continent_Oceania <= 0.50
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 16.00] class: 1
|   |   |   |   |   |   |   |   |--- continent_Oceania >  0.50
|   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- prevailing_wage >  7522.50
|   |   |   |   |   |   |   |   |--- region_of_employment_Midwest <= 0.50
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 612.00
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 335.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 6
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  335.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  612.00
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 909.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  909.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 20
|   |   |   |   |   |   |   |   |--- region_of_employment_Midwest >  0.50
|   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 2011.50
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 32908.69
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  32908.69
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |   |   |--- yr_of_estab >  2011.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 5463.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [4.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  5463.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |--- continent_Asia >  0.50
|   |   |   |   |   |   |   |--- no_of_employees <= 24.00
|   |   |   |   |   |   |   |   |--- weights: [5.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- no_of_employees >  24.00
|   |   |   |   |   |   |   |   |--- prevailing_wage <= 188467.90
|   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 1515.91
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 2003.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab >  2003.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 4.00] class: 1
|   |   |   |   |   |   |   |   |   |--- prevailing_wage >  1515.91
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 2424.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 39
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  2424.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 27
|   |   |   |   |   |   |   |   |--- prevailing_wage >  188467.90
|   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 2004.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 245.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  245.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 9
|   |   |   |   |   |   |   |   |   |--- yr_of_estab >  2004.50
|   |   |   |   |   |   |   |   |   |   |--- region_of_employment_West <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |   |--- region_of_employment_West >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |--- continent_Europe >  0.50
|   |   |   |   |   |   |--- prevailing_wage <= 158681.25
|   |   |   |   |   |   |   |--- yr_of_estab <= 1987.50
|   |   |   |   |   |   |   |   |--- prevailing_wage <= 125827.25
|   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 38093.38
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 241.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  241.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 7
|   |   |   |   |   |   |   |   |   |--- prevailing_wage >  38093.38
|   |   |   |   |   |   |   |   |   |   |--- region_of_employment_South <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 15
|   |   |   |   |   |   |   |   |   |   |--- region_of_employment_South >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 7
|   |   |   |   |   |   |   |   |--- prevailing_wage >  125827.25
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 17.00] class: 1
|   |   |   |   |   |   |   |--- yr_of_estab >  1987.50
|   |   |   |   |   |   |   |   |--- prevailing_wage <= 130823.08
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 1804.50
|   |   |   |   |   |   |   |   |   |   |--- region_of_employment_Midwest <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 7
|   |   |   |   |   |   |   |   |   |   |--- region_of_employment_Midwest >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  1804.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 1816.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  1816.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 13
|   |   |   |   |   |   |   |   |--- prevailing_wage >  130823.08
|   |   |   |   |   |   |   |   |   |--- region_of_employment_South <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 2011.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab >  2011.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- region_of_employment_South >  0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 4.00] class: 1
|   |   |   |   |   |   |--- prevailing_wage >  158681.25
|   |   |   |   |   |   |   |--- prevailing_wage <= 216135.16
|   |   |   |   |   |   |   |   |--- region_of_employment_South <= 0.50
|   |   |   |   |   |   |   |   |   |--- region_of_employment_Northeast <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 167895.16
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  167895.16
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [5.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- region_of_employment_Northeast >  0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |--- region_of_employment_South >  0.50
|   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 165582.23
|   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- prevailing_wage >  165582.23
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 3.00] class: 1
|   |   |   |   |   |   |   |--- prevailing_wage >  216135.16
|   |   |   |   |   |   |   |   |--- full_time_position <= 0.50
|   |   |   |   |   |   |   |   |   |--- requires_job_training <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- requires_job_training >  0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |--- full_time_position >  0.50
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 5.00] class: 1
|   |   |   |--- education_of_employee_Doctorate >  0.50
|   |   |   |   |--- unit_of_wage_Year <= 0.50
|   |   |   |   |   |--- yr_of_estab <= 2006.50
|   |   |   |   |   |   |--- no_of_employees <= 2536.00
|   |   |   |   |   |   |   |--- yr_of_estab <= 1997.50
|   |   |   |   |   |   |   |   |--- yr_of_estab <= 1964.00
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 3.00] class: 1
|   |   |   |   |   |   |   |   |--- yr_of_estab >  1964.00
|   |   |   |   |   |   |   |   |   |--- unit_of_wage_Week <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- unit_of_wage_Week >  0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |--- yr_of_estab >  1997.50
|   |   |   |   |   |   |   |   |--- weights: [0.00, 5.00] class: 1
|   |   |   |   |   |   |--- no_of_employees >  2536.00
|   |   |   |   |   |   |   |--- region_of_employment_South <= 0.50
|   |   |   |   |   |   |   |   |--- yr_of_estab <= 1996.00
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 4.00] class: 1
|   |   |   |   |   |   |   |   |--- yr_of_estab >  1996.00
|   |   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- region_of_employment_South >  0.50
|   |   |   |   |   |   |   |   |--- weights: [3.00, 0.00] class: 0
|   |   |   |   |   |--- yr_of_estab >  2006.50
|   |   |   |   |   |   |--- weights: [0.00, 7.00] class: 1
|   |   |   |   |--- unit_of_wage_Year >  0.50
|   |   |   |   |   |--- requires_job_training <= 0.50
|   |   |   |   |   |   |--- continent_North America <= 0.50
|   |   |   |   |   |   |   |--- region_of_employment_South <= 0.50
|   |   |   |   |   |   |   |   |--- region_of_employment_Midwest <= 0.50
|   |   |   |   |   |   |   |   |   |--- continent_Oceania <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- continent_South America <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 14
|   |   |   |   |   |   |   |   |   |   |--- continent_South America >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |   |--- continent_Oceania >  0.50
|   |   |   |   |   |   |   |   |   |   |--- region_of_employment_Northeast <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- region_of_employment_Northeast >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |   |   |   |--- region_of_employment_Midwest >  0.50
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 69.00] class: 1
|   |   |   |   |   |   |   |--- region_of_employment_South >  0.50
|   |   |   |   |   |   |   |   |--- full_time_position <= 0.50
|   |   |   |   |   |   |   |   |   |--- continent_Asia <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- continent_Asia >  0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 8.00] class: 1
|   |   |   |   |   |   |   |   |--- full_time_position >  0.50
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 147.00] class: 1
|   |   |   |   |   |   |--- continent_North America >  0.50
|   |   |   |   |   |   |   |--- no_of_employees <= 242.50
|   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- no_of_employees >  242.50
|   |   |   |   |   |   |   |   |--- yr_of_estab <= 2002.50
|   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 107494.71
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 103225.95
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 10
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  103225.95
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- prevailing_wage >  107494.71
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 15.00] class: 1
|   |   |   |   |   |   |   |   |--- yr_of_estab >  2002.50
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 23.00] class: 1
|   |   |   |   |   |--- requires_job_training >  0.50
|   |   |   |   |   |   |--- yr_of_estab <= 2013.50
|   |   |   |   |   |   |   |--- region_of_employment_Northeast <= 0.50
|   |   |   |   |   |   |   |   |--- no_of_employees <= 7072.50
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 285.00
|   |   |   |   |   |   |   |   |   |   |--- continent_Europe <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 7.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- continent_Europe >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  285.00
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 140409.80
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  140409.80
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |--- no_of_employees >  7072.50
|   |   |   |   |   |   |   |   |   |--- continent_Europe <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- region_of_employment_Midwest <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- region_of_employment_Midwest >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |--- continent_Europe >  0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |   |   |--- region_of_employment_Northeast >  0.50
|   |   |   |   |   |   |   |   |--- prevailing_wage <= 80474.88
|   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 76966.84
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 1982.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 5.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab >  1982.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |   |   |--- prevailing_wage >  76966.84
|   |   |   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- prevailing_wage >  80474.88
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 10.00] class: 1
|   |   |   |   |   |   |--- yr_of_estab >  2013.50
|   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |--- education_of_employee_Master's >  0.50
|   |   |   |--- unit_of_wage_Year <= 0.50
|   |   |   |   |--- continent_Asia <= 0.50
|   |   |   |   |   |--- continent_Europe <= 0.50
|   |   |   |   |   |   |--- unit_of_wage_Week <= 0.50
|   |   |   |   |   |   |   |--- no_of_employees <= 3481.00
|   |   |   |   |   |   |   |   |--- no_of_employees <= 1440.50
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 341.50
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 2011.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab >  2011.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  341.50
|   |   |   |   |   |   |   |   |   |   |--- region_of_employment_West <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |   |--- region_of_employment_West >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |--- no_of_employees >  1440.50
|   |   |   |   |   |   |   |   |   |--- region_of_employment_West <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 2008.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [15.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab >  2008.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |--- region_of_employment_West >  0.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 3367.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 3.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  3367.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- no_of_employees >  3481.00
|   |   |   |   |   |   |   |   |--- yr_of_estab <= 1940.75
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 3758.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  3758.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- yr_of_estab >  1940.75
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 6.00] class: 1
|   |   |   |   |   |   |--- unit_of_wage_Week >  0.50
|   |   |   |   |   |   |   |--- no_of_employees <= 4385.00
|   |   |   |   |   |   |   |   |--- no_of_employees <= 594.00
|   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- no_of_employees >  594.00
|   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 2004.00
|   |   |   |   |   |   |   |   |   |   |--- region_of_employment_West <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 10.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- region_of_employment_West >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |--- yr_of_estab >  2004.00
|   |   |   |   |   |   |   |   |   |   |--- continent_North America <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- continent_North America >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |--- no_of_employees >  4385.00
|   |   |   |   |   |   |   |   |--- region_of_employment_Northeast <= 0.50
|   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- region_of_employment_Northeast >  0.50
|   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 1954.25
|   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- yr_of_estab >  1954.25
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |--- continent_Europe >  0.50
|   |   |   |   |   |   |--- full_time_position <= 0.50
|   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |--- full_time_position >  0.50
|   |   |   |   |   |   |   |--- requires_job_training <= 0.50
|   |   |   |   |   |   |   |   |--- weights: [0.00, 14.00] class: 1
|   |   |   |   |   |   |   |--- requires_job_training >  0.50
|   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |--- continent_Asia >  0.50
|   |   |   |   |   |--- no_of_employees <= 84.00
|   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |--- no_of_employees >  84.00
|   |   |   |   |   |   |--- no_of_employees <= 2608.00
|   |   |   |   |   |   |   |--- requires_job_training <= 0.50
|   |   |   |   |   |   |   |   |--- no_of_employees <= 469.00
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 16.00] class: 1
|   |   |   |   |   |   |   |   |--- no_of_employees >  469.00
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 478.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  478.50
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 209973.79
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  209973.79
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 11
|   |   |   |   |   |   |   |--- requires_job_training >  0.50
|   |   |   |   |   |   |   |   |--- yr_of_estab <= 1987.50
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |--- yr_of_estab >  1987.50
|   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |--- no_of_employees >  2608.00
|   |   |   |   |   |   |   |--- no_of_employees <= 2659.50
|   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- no_of_employees >  2659.50
|   |   |   |   |   |   |   |   |--- full_time_position <= 0.50
|   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- full_time_position >  0.50
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 6324.00
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 4449.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 9
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  4449.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  6324.00
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 6.00] class: 1
|   |   |   |--- unit_of_wage_Year >  0.50
|   |   |   |   |--- region_of_employment_West <= 0.50
|   |   |   |   |   |--- full_time_position <= 0.50
|   |   |   |   |   |   |--- no_of_employees <= 132.50
|   |   |   |   |   |   |   |--- region_of_employment_Northeast <= 0.50
|   |   |   |   |   |   |   |   |--- no_of_employees <= 58.50
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 4.00] class: 1
|   |   |   |   |   |   |   |   |--- no_of_employees >  58.50
|   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 134699.42
|   |   |   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- prevailing_wage >  134699.42
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |--- region_of_employment_Northeast >  0.50
|   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |--- no_of_employees >  132.50
|   |   |   |   |   |   |   |--- prevailing_wage <= 4850.04
|   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- prevailing_wage >  4850.04
|   |   |   |   |   |   |   |   |--- region_of_employment_Midwest <= 0.50
|   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 165227.40
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 149482.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 13
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  149482.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |--- prevailing_wage >  165227.40
|   |   |   |   |   |   |   |   |   |   |--- continent_North America <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 8
|   |   |   |   |   |   |   |   |   |   |--- continent_North America >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |--- region_of_employment_Midwest >  0.50
|   |   |   |   |   |   |   |   |   |--- continent_Oceania <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 113653.92
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 8
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  113653.92
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |   |   |--- continent_Oceania >  0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |--- full_time_position >  0.50
|   |   |   |   |   |   |--- region_of_employment_Northeast <= 0.50
|   |   |   |   |   |   |   |--- requires_job_training <= 0.50
|   |   |   |   |   |   |   |   |--- continent_South America <= 0.50
|   |   |   |   |   |   |   |   |   |--- continent_North America <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- continent_Oceania <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 7
|   |   |   |   |   |   |   |   |   |   |--- continent_Oceania >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |--- continent_North America >  0.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 7060.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 7
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  7060.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |--- continent_South America >  0.50
|   |   |   |   |   |   |   |   |   |--- region_of_employment_South <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- region_of_employment_Midwest <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- region_of_employment_Midwest >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 10.00] class: 1
|   |   |   |   |   |   |   |   |   |--- region_of_employment_South >  0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 17.00] class: 1
|   |   |   |   |   |   |   |--- requires_job_training >  0.50
|   |   |   |   |   |   |   |   |--- prevailing_wage <= 228499.93
|   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 2011.00
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 1463.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 33.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  1463.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 8
|   |   |   |   |   |   |   |   |   |--- yr_of_estab >  2011.00
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 239.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  239.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |--- prevailing_wage >  228499.93
|   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |--- region_of_employment_Northeast >  0.50
|   |   |   |   |   |   |   |--- continent_North America <= 0.50
|   |   |   |   |   |   |   |   |--- requires_job_training <= 0.50
|   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 2000.50
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 1999.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 20
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab >  1999.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 6
|   |   |   |   |   |   |   |   |   |--- yr_of_estab >  2000.50
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 509.51
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  509.51
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 14
|   |   |   |   |   |   |   |   |--- requires_job_training >  0.50
|   |   |   |   |   |   |   |   |   |--- continent_Oceania <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 51707.83
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  51707.83
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 6
|   |   |   |   |   |   |   |   |   |--- continent_Oceania >  0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- continent_North America >  0.50
|   |   |   |   |   |   |   |   |--- no_of_employees <= 114.50
|   |   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- no_of_employees >  114.50
|   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 1968.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 6551.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 9
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  6551.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- yr_of_estab >  1968.50
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 1094.35
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  1094.35
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 12
|   |   |   |   |--- region_of_employment_West >  0.50
|   |   |   |   |   |--- continent_South America <= 0.50
|   |   |   |   |   |   |--- continent_North America <= 0.50
|   |   |   |   |   |   |   |--- continent_Oceania <= 0.50
|   |   |   |   |   |   |   |   |--- continent_Asia <= 0.50
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 96.00
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 1997.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [3.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab >  1997.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  96.00
|   |   |   |   |   |   |   |   |   |   |--- full_time_position <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- full_time_position >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 8
|   |   |   |   |   |   |   |   |--- continent_Asia >  0.50
|   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 2001.50
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 1997.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 16
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab >  1997.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 11
|   |   |   |   |   |   |   |   |   |--- yr_of_estab >  2001.50
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 16170.72
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 7
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  16170.72
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 13
|   |   |   |   |   |   |   |--- continent_Oceania >  0.50
|   |   |   |   |   |   |   |   |--- no_of_employees <= 2123.00
|   |   |   |   |   |   |   |   |   |--- full_time_position <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- full_time_position >  0.50
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 55312.33
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  55312.33
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |--- no_of_employees >  2123.00
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 3.00] class: 1
|   |   |   |   |   |   |--- continent_North America >  0.50
|   |   |   |   |   |   |   |--- yr_of_estab <= 2011.50
|   |   |   |   |   |   |   |   |--- prevailing_wage <= 10972.25
|   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 1962.00
|   |   |   |   |   |   |   |   |   |   |--- weights: [3.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- yr_of_estab >  1962.00
|   |   |   |   |   |   |   |   |   |   |--- full_time_position <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- full_time_position >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |--- prevailing_wage >  10972.25
|   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 79129.82
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 1663.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 14.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  1663.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 10
|   |   |   |   |   |   |   |   |   |--- prevailing_wage >  79129.82
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 131951.44
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 8
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  131951.44
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |--- yr_of_estab >  2011.50
|   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |--- continent_South America >  0.50
|   |   |   |   |   |   |--- prevailing_wage <= 32867.22
|   |   |   |   |   |   |   |--- yr_of_estab <= 1998.50
|   |   |   |   |   |   |   |   |--- weights: [0.00, 5.00] class: 1
|   |   |   |   |   |   |   |--- yr_of_estab >  1998.50
|   |   |   |   |   |   |   |   |--- yr_of_estab <= 2004.00
|   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- yr_of_estab >  2004.00
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 3.00] class: 1
|   |   |   |   |   |   |--- prevailing_wage >  32867.22
|   |   |   |   |   |   |   |--- yr_of_estab <= 1998.50
|   |   |   |   |   |   |   |   |--- prevailing_wage <= 113576.45
|   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 59437.28
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 1648.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  1648.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |--- prevailing_wage >  59437.28
|   |   |   |   |   |   |   |   |   |   |--- weights: [3.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- prevailing_wage >  113576.45
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 3.00] class: 1
|   |   |   |   |   |   |   |--- yr_of_estab >  1998.50
|   |   |   |   |   |   |   |   |--- weights: [6.00, 0.00] class: 0
|--- education_of_employee_High School >  0.50
|   |--- continent_Asia <= 0.50
|   |   |--- continent_Europe <= 0.50
|   |   |   |--- has_job_experience <= 0.50
|   |   |   |   |--- yr_of_estab <= 1968.00
|   |   |   |   |   |--- yr_of_estab <= 1961.50
|   |   |   |   |   |   |--- yr_of_estab <= 1946.50
|   |   |   |   |   |   |   |--- prevailing_wage <= 46697.76
|   |   |   |   |   |   |   |   |--- weights: [0.00, 5.00] class: 1
|   |   |   |   |   |   |   |--- prevailing_wage >  46697.76
|   |   |   |   |   |   |   |   |--- region_of_employment_South <= 0.50
|   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 57871.71
|   |   |   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- prevailing_wage >  57871.71
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 5129.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  5129.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- region_of_employment_South >  0.50
|   |   |   |   |   |   |   |   |   |--- weights: [4.00, 0.00] class: 0
|   |   |   |   |   |   |--- yr_of_estab >  1946.50
|   |   |   |   |   |   |   |--- requires_job_training <= 0.50
|   |   |   |   |   |   |   |   |--- weights: [4.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- requires_job_training >  0.50
|   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |--- yr_of_estab >  1961.50
|   |   |   |   |   |   |--- weights: [0.00, 4.00] class: 1
|   |   |   |   |--- yr_of_estab >  1968.00
|   |   |   |   |   |--- region_of_employment_Northeast <= 0.50
|   |   |   |   |   |   |--- yr_of_estab <= 2009.50
|   |   |   |   |   |   |   |--- yr_of_estab <= 2006.50
|   |   |   |   |   |   |   |   |--- no_of_employees <= 2553.50
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 1365.50
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 1996.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab >  1996.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  1365.50
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 1976.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab >  1976.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |--- no_of_employees >  2553.50
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 2666.00
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 3.00] class: 1
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  2666.00
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 1997.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab >  1997.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |--- yr_of_estab >  2006.50
|   |   |   |   |   |   |   |   |--- continent_Oceania <= 0.50
|   |   |   |   |   |   |   |   |   |--- weights: [9.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- continent_Oceania >  0.50
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 767.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  767.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |--- yr_of_estab >  2009.50
|   |   |   |   |   |   |   |--- region_of_employment_South <= 0.50
|   |   |   |   |   |   |   |   |--- yr_of_estab <= 2012.00
|   |   |   |   |   |   |   |   |   |--- unit_of_wage_Year <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- unit_of_wage_Year >  0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |   |   |   |--- yr_of_estab >  2012.00
|   |   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- region_of_employment_South >  0.50
|   |   |   |   |   |   |   |   |--- weights: [0.00, 3.00] class: 1
|   |   |   |   |   |--- region_of_employment_Northeast >  0.50
|   |   |   |   |   |   |--- no_of_employees <= 6664.50
|   |   |   |   |   |   |   |--- no_of_employees <= 3089.00
|   |   |   |   |   |   |   |   |--- no_of_employees <= 1365.00
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 1248.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [11.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  1248.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |--- no_of_employees >  1365.00
|   |   |   |   |   |   |   |   |   |--- weights: [23.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- no_of_employees >  3089.00
|   |   |   |   |   |   |   |   |--- prevailing_wage <= 90632.39
|   |   |   |   |   |   |   |   |   |--- weights: [4.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- prevailing_wage >  90632.39
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 3927.00
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  3927.00
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 116871.55
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  116871.55
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |--- no_of_employees >  6664.50
|   |   |   |   |   |   |   |--- prevailing_wage <= 141638.03
|   |   |   |   |   |   |   |   |--- yr_of_estab <= 1993.00
|   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- yr_of_estab >  1993.00
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |--- prevailing_wage >  141638.03
|   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |--- has_job_experience >  0.50
|   |   |   |   |--- prevailing_wage <= 21595.07
|   |   |   |   |   |--- unit_of_wage_Year <= 0.50
|   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |--- unit_of_wage_Year >  0.50
|   |   |   |   |   |   |--- continent_South America <= 0.50
|   |   |   |   |   |   |   |--- no_of_employees <= 6480.50
|   |   |   |   |   |   |   |   |--- weights: [0.00, 24.00] class: 1
|   |   |   |   |   |   |   |--- no_of_employees >  6480.50
|   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |--- continent_South America >  0.50
|   |   |   |   |   |   |   |--- yr_of_estab <= 1998.50
|   |   |   |   |   |   |   |   |--- no_of_employees <= 3115.00
|   |   |   |   |   |   |   |   |   |--- weights: [3.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- no_of_employees >  3115.00
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |--- yr_of_estab >  1998.50
|   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |--- prevailing_wage >  21595.07
|   |   |   |   |   |--- prevailing_wage <= 24023.89
|   |   |   |   |   |   |--- no_of_employees <= 1241.00
|   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |--- no_of_employees >  1241.00
|   |   |   |   |   |   |   |--- weights: [4.00, 0.00] class: 0
|   |   |   |   |   |--- prevailing_wage >  24023.89
|   |   |   |   |   |   |--- no_of_employees <= 473.00
|   |   |   |   |   |   |   |--- prevailing_wage <= 160790.63
|   |   |   |   |   |   |   |   |--- continent_North America <= 0.50
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 228.00
|   |   |   |   |   |   |   |   |   |   |--- weights: [4.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  228.00
|   |   |   |   |   |   |   |   |   |   |--- region_of_employment_Midwest <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- region_of_employment_Midwest >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |--- continent_North America >  0.50
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 185.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 8.00] class: 1
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  185.50
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 2009.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab >  2009.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [3.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- prevailing_wage >  160790.63
|   |   |   |   |   |   |   |   |--- weights: [5.00, 0.00] class: 0
|   |   |   |   |   |   |--- no_of_employees >  473.00
|   |   |   |   |   |   |   |--- no_of_employees <= 981.50
|   |   |   |   |   |   |   |   |--- yr_of_estab <= 1995.00
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 671.00
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 545.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  545.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [3.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  671.00
|   |   |   |   |   |   |   |   |   |   |--- region_of_employment_South <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 6.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- region_of_employment_South >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- yr_of_estab >  1995.00
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 16.00] class: 1
|   |   |   |   |   |   |   |--- no_of_employees >  981.50
|   |   |   |   |   |   |   |   |--- yr_of_estab <= 2009.00
|   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 126562.11
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 1946.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab >  1946.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 12
|   |   |   |   |   |   |   |   |   |--- prevailing_wage >  126562.11
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 2743.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 8
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  2743.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |--- yr_of_estab >  2009.00
|   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 2013.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 2544.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 8.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  2544.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |--- yr_of_estab >  2013.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 3118.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  3118.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |--- continent_Europe >  0.50
|   |   |   |--- region_of_employment_West <= 0.50
|   |   |   |   |--- prevailing_wage <= 4025.76
|   |   |   |   |   |--- weights: [5.00, 0.00] class: 0
|   |   |   |   |--- prevailing_wage >  4025.76
|   |   |   |   |   |--- prevailing_wage <= 5798.96
|   |   |   |   |   |   |--- weights: [0.00, 3.00] class: 1
|   |   |   |   |   |--- prevailing_wage >  5798.96
|   |   |   |   |   |   |--- yr_of_estab <= 1935.00
|   |   |   |   |   |   |   |--- no_of_employees <= 125.50
|   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |   |   |--- no_of_employees >  125.50
|   |   |   |   |   |   |   |   |--- prevailing_wage <= 142231.16
|   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 37092.47
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 14228.38
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [4.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  14228.38
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |--- prevailing_wage >  37092.47
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 308.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  308.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |--- prevailing_wage >  142231.16
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 2656.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  2656.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |--- yr_of_estab >  1935.00
|   |   |   |   |   |   |   |--- yr_of_estab <= 1966.50
|   |   |   |   |   |   |   |   |--- requires_job_training <= 0.50
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 5301.00
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 3015.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 9.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  3015.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  5301.00
|   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- requires_job_training >  0.50
|   |   |   |   |   |   |   |   |   |--- weights: [3.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- yr_of_estab >  1966.50
|   |   |   |   |   |   |   |   |--- prevailing_wage <= 46240.11
|   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 1992.50
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 1976.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab >  1976.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |   |--- yr_of_estab >  1992.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 916.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  916.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 7
|   |   |   |   |   |   |   |   |--- prevailing_wage >  46240.11
|   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 1985.50
|   |   |   |   |   |   |   |   |   |   |--- region_of_employment_Midwest <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |   |--- region_of_employment_Midwest >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |   |   |   |   |--- yr_of_estab >  1985.50
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 52136.59
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  52136.59
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 15
|   |   |   |--- region_of_employment_West >  0.50
|   |   |   |   |--- no_of_employees <= 3180.00
|   |   |   |   |   |--- prevailing_wage <= 2497.47
|   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |--- prevailing_wage >  2497.47
|   |   |   |   |   |   |--- no_of_employees <= 64.50
|   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |--- no_of_employees >  64.50
|   |   |   |   |   |   |   |--- yr_of_estab <= 2006.50
|   |   |   |   |   |   |   |   |--- yr_of_estab <= 2005.50
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 479.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [12.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  479.50
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 97411.21
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 9
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  97411.21
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [11.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- yr_of_estab >  2005.50
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |--- yr_of_estab >  2006.50
|   |   |   |   |   |   |   |   |--- weights: [15.00, 0.00] class: 0
|   |   |   |   |--- no_of_employees >  3180.00
|   |   |   |   |   |--- no_of_employees <= 3331.50
|   |   |   |   |   |   |--- weights: [0.00, 3.00] class: 1
|   |   |   |   |   |--- no_of_employees >  3331.50
|   |   |   |   |   |   |--- yr_of_estab <= 1975.50
|   |   |   |   |   |   |   |--- yr_of_estab <= 1954.50
|   |   |   |   |   |   |   |   |--- no_of_employees <= 5479.00
|   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 38981.93
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 18602.95
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  18602.95
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- prevailing_wage >  38981.93
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |   |   |   |--- no_of_employees >  5479.00
|   |   |   |   |   |   |   |   |   |--- weights: [3.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- yr_of_estab >  1954.50
|   |   |   |   |   |   |   |   |--- weights: [0.00, 4.00] class: 1
|   |   |   |   |   |   |--- yr_of_estab >  1975.50
|   |   |   |   |   |   |   |--- yr_of_estab <= 2011.00
|   |   |   |   |   |   |   |   |--- no_of_employees <= 5369.50
|   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 2004.00
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 83756.20
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [14.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  83756.20
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |--- yr_of_estab >  2004.00
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 113804.24
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  113804.24
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- no_of_employees >  5369.50
|   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 1993.00
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |   |   |   |   |--- yr_of_estab >  1993.00
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 36883.51
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [3.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  36883.51
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |--- yr_of_estab >  2011.00
|   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |--- continent_Asia >  0.50
|   |   |--- region_of_employment_West <= 0.50
|   |   |   |--- region_of_employment_Northeast <= 0.50
|   |   |   |   |--- prevailing_wage <= 5379.81
|   |   |   |   |   |--- weights: [10.00, 0.00] class: 0
|   |   |   |   |--- prevailing_wage >  5379.81
|   |   |   |   |   |--- yr_of_estab <= 1996.50
|   |   |   |   |   |   |--- yr_of_estab <= 1983.50
|   |   |   |   |   |   |   |--- no_of_employees <= 312.50
|   |   |   |   |   |   |   |   |--- no_of_employees <= 81.50
|   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 67099.56
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |--- prevailing_wage >  67099.56
|   |   |   |   |   |   |   |   |   |   |--- weights: [5.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- no_of_employees >  81.50
|   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 95508.68
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 50513.11
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 3.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  50513.11
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [5.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- prevailing_wage >  95508.68
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 9.00] class: 1
|   |   |   |   |   |   |   |--- no_of_employees >  312.50
|   |   |   |   |   |   |   |   |--- no_of_employees <= 935.00
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 699.00
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 611.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 6
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  611.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  699.00
|   |   |   |   |   |   |   |   |   |   |--- weights: [12.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- no_of_employees >  935.00
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 1278.50
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 1971.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab >  1971.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [3.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  1278.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 1327.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [3.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  1327.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 22
|   |   |   |   |   |   |--- yr_of_estab >  1983.50
|   |   |   |   |   |   |   |--- no_of_employees <= 1616.50
|   |   |   |   |   |   |   |   |--- no_of_employees <= 28.00
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |--- no_of_employees >  28.00
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 254.00
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 189.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [3.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  189.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  254.00
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 32000.36
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  32000.36
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 6
|   |   |   |   |   |   |   |--- no_of_employees >  1616.50
|   |   |   |   |   |   |   |   |--- yr_of_estab <= 1992.50
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 5357.00
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 55748.05
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  55748.05
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 8
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  5357.00
|   |   |   |   |   |   |   |   |   |   |--- requires_job_training <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 2.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- requires_job_training >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- yr_of_estab >  1992.50
|   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 133422.91
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 1760.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  1760.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 6
|   |   |   |   |   |   |   |   |   |--- prevailing_wage >  133422.91
|   |   |   |   |   |   |   |   |   |   |--- unit_of_wage_Year <= 0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- unit_of_wage_Year >  0.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |--- yr_of_estab >  1996.50
|   |   |   |   |   |   |--- no_of_employees <= 588.50
|   |   |   |   |   |   |   |--- requires_job_training <= 0.50
|   |   |   |   |   |   |   |   |--- prevailing_wage <= 19372.01
|   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 12600.46
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 6978.48
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  6978.48
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- prevailing_wage >  12600.46
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 3.00] class: 1
|   |   |   |   |   |   |   |   |--- prevailing_wage >  19372.01
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 505.00
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 495.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 9
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  495.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  505.00
|   |   |   |   |   |   |   |   |   |   |--- weights: [10.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- requires_job_training >  0.50
|   |   |   |   |   |   |   |   |--- prevailing_wage <= 83324.25
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 365.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  365.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |--- prevailing_wage >  83324.25
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 3.00] class: 1
|   |   |   |   |   |   |--- no_of_employees >  588.50
|   |   |   |   |   |   |   |--- no_of_employees <= 629.00
|   |   |   |   |   |   |   |   |--- yr_of_estab <= 1997.50
|   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- yr_of_estab >  1997.50
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 5.00] class: 1
|   |   |   |   |   |   |   |--- no_of_employees >  629.00
|   |   |   |   |   |   |   |   |--- has_job_experience <= 0.50
|   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 66641.22
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 2000.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab >  2000.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |   |   |--- prevailing_wage >  66641.22
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 87682.91
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 8
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  87682.91
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 14
|   |   |   |   |   |   |   |   |--- has_job_experience >  0.50
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 5267.00
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 4562.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 16
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  4562.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  5267.00
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 2001.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab >  2001.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [13.00, 0.00] class: 0
|   |   |   |--- region_of_employment_Northeast >  0.50
|   |   |   |   |--- full_time_position <= 0.50
|   |   |   |   |   |--- weights: [32.00, 0.00] class: 0
|   |   |   |   |--- full_time_position >  0.50
|   |   |   |   |   |--- unit_of_wage_Year <= 0.50
|   |   |   |   |   |   |--- requires_job_training <= 0.50
|   |   |   |   |   |   |   |--- yr_of_estab <= 1999.50
|   |   |   |   |   |   |   |   |--- yr_of_estab <= 1977.50
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 546.00
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 261.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  261.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  546.00
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 1975.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [10.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab >  1975.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |--- yr_of_estab >  1977.50
|   |   |   |   |   |   |   |   |   |--- weights: [25.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- yr_of_estab >  1999.50
|   |   |   |   |   |   |   |   |--- prevailing_wage <= 94310.40
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |--- prevailing_wage >  94310.40
|   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 2002.00
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 2761.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  2761.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- yr_of_estab >  2002.00
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 2012.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [13.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab >  2012.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |--- requires_job_training >  0.50
|   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |--- unit_of_wage_Year >  0.50
|   |   |   |   |   |   |--- no_of_employees <= 139.50
|   |   |   |   |   |   |   |--- yr_of_estab <= 2011.00
|   |   |   |   |   |   |   |   |--- weights: [15.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- yr_of_estab >  2011.00
|   |   |   |   |   |   |   |   |--- prevailing_wage <= 51519.17
|   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- prevailing_wage >  51519.17
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |--- no_of_employees >  139.50
|   |   |   |   |   |   |   |--- prevailing_wage <= 44895.46
|   |   |   |   |   |   |   |   |--- no_of_employees <= 277.00
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |--- no_of_employees >  277.00
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 3014.00
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 2068.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 6
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  2068.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  3014.00
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 1979.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [15.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- yr_of_estab >  1979.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 4
|   |   |   |   |   |   |   |--- prevailing_wage >  44895.46
|   |   |   |   |   |   |   |   |--- prevailing_wage <= 87325.16
|   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 2008.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 591.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  591.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 15
|   |   |   |   |   |   |   |   |   |--- yr_of_estab >  2008.50
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 85202.47
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 3
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  85202.47
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |--- prevailing_wage >  87325.16
|   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 143328.84
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 172.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  172.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 9
|   |   |   |   |   |   |   |   |   |--- prevailing_wage >  143328.84
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 2869.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 6
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  2869.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 5
|   |   |--- region_of_employment_West >  0.50
|   |   |   |--- no_of_employees <= 5548.50
|   |   |   |   |--- no_of_employees <= 512.50
|   |   |   |   |   |--- prevailing_wage <= 108693.75
|   |   |   |   |   |   |--- full_time_position <= 0.50
|   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |--- full_time_position >  0.50
|   |   |   |   |   |   |   |--- prevailing_wage <= 14820.68
|   |   |   |   |   |   |   |   |--- prevailing_wage <= 11777.77
|   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- prevailing_wage >  11777.77
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |--- prevailing_wage >  14820.68
|   |   |   |   |   |   |   |   |--- yr_of_estab <= 2010.00
|   |   |   |   |   |   |   |   |   |--- yr_of_estab <= 1970.00
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 228.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  228.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- yr_of_estab >  1970.00
|   |   |   |   |   |   |   |   |   |   |--- weights: [15.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- yr_of_estab >  2010.00
|   |   |   |   |   |   |   |   |   |--- requires_job_training <= 0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |--- requires_job_training >  0.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |--- prevailing_wage >  108693.75
|   |   |   |   |   |   |--- yr_of_estab <= 2013.00
|   |   |   |   |   |   |   |--- no_of_employees <= 98.50
|   |   |   |   |   |   |   |   |--- has_job_experience <= 0.50
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |--- has_job_experience >  0.50
|   |   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |   |   |--- no_of_employees >  98.50
|   |   |   |   |   |   |   |   |--- yr_of_estab <= 2004.00
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 5.00] class: 1
|   |   |   |   |   |   |   |   |--- yr_of_estab >  2004.00
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 479.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  479.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |--- yr_of_estab >  2013.00
|   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |--- no_of_employees >  512.50
|   |   |   |   |   |--- yr_of_estab <= 2011.50
|   |   |   |   |   |   |--- yr_of_estab <= 1933.75
|   |   |   |   |   |   |   |--- no_of_employees <= 1003.00
|   |   |   |   |   |   |   |   |--- no_of_employees <= 957.50
|   |   |   |   |   |   |   |   |   |--- weights: [7.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- no_of_employees >  957.50
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |--- no_of_employees >  1003.00
|   |   |   |   |   |   |   |   |--- weights: [28.00, 0.00] class: 0
|   |   |   |   |   |   |--- yr_of_estab >  1933.75
|   |   |   |   |   |   |   |--- yr_of_estab <= 1936.00
|   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |--- yr_of_estab >  1936.00
|   |   |   |   |   |   |   |   |--- no_of_employees <= 1293.00
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 884.50
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees <= 805.50
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [11.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- no_of_employees >  805.50
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 2
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  884.50
|   |   |   |   |   |   |   |   |   |   |--- weights: [23.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |--- no_of_employees >  1293.00
|   |   |   |   |   |   |   |   |   |--- no_of_employees <= 1319.00
|   |   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |   |--- no_of_employees >  1319.00
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage <= 11372.00
|   |   |   |   |   |   |   |   |   |   |   |--- weights: [13.00, 0.00] class: 0
|   |   |   |   |   |   |   |   |   |   |--- prevailing_wage >  11372.00
|   |   |   |   |   |   |   |   |   |   |   |--- truncated branch of depth 13
|   |   |   |   |   |--- yr_of_estab >  2011.50
|   |   |   |   |   |   |--- no_of_employees <= 1382.50
|   |   |   |   |   |   |   |--- yr_of_estab <= 2012.50
|   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |--- yr_of_estab >  2012.50
|   |   |   |   |   |   |   |   |--- weights: [4.00, 0.00] class: 0
|   |   |   |   |   |   |--- no_of_employees >  1382.50
|   |   |   |   |   |   |   |--- no_of_employees <= 2889.00
|   |   |   |   |   |   |   |   |--- weights: [0.00, 3.00] class: 1
|   |   |   |   |   |   |   |--- no_of_employees >  2889.00
|   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |--- no_of_employees >  5548.50
|   |   |   |   |--- no_of_employees <= 6854.00
|   |   |   |   |   |--- yr_of_estab <= 2005.00
|   |   |   |   |   |   |--- weights: [0.00, 5.00] class: 1
|   |   |   |   |   |--- yr_of_estab >  2005.00
|   |   |   |   |   |   |--- prevailing_wage <= 28867.76
|   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |--- prevailing_wage >  28867.76
|   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |--- no_of_employees >  6854.00
|   |   |   |   |   |--- yr_of_estab <= 1989.50
|   |   |   |   |   |   |--- yr_of_estab <= 1971.50
|   |   |   |   |   |   |   |--- weights: [3.00, 0.00] class: 0
|   |   |   |   |   |   |--- yr_of_estab >  1971.50
|   |   |   |   |   |   |   |--- prevailing_wage <= 92896.77
|   |   |   |   |   |   |   |   |--- weights: [0.00, 3.00] class: 1
|   |   |   |   |   |   |   |--- prevailing_wage >  92896.77
|   |   |   |   |   |   |   |   |--- requires_job_training <= 0.50
|   |   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |   |--- requires_job_training >  0.50
|   |   |   |   |   |   |   |   |   |--- weights: [2.00, 0.00] class: 0
|   |   |   |   |   |--- yr_of_estab >  1989.50
|   |   |   |   |   |   |--- prevailing_wage <= 56150.82
|   |   |   |   |   |   |   |--- yr_of_estab <= 2002.00
|   |   |   |   |   |   |   |   |--- weights: [0.00, 1.00] class: 1
|   |   |   |   |   |   |   |--- yr_of_estab >  2002.00
|   |   |   |   |   |   |   |   |--- weights: [1.00, 0.00] class: 0
|   |   |   |   |   |   |--- prevailing_wage >  56150.82
|   |   |   |   |   |   |   |--- weights: [9.00, 0.00] class: 0

importance_of_attributes = pd.DataFrame(dTree.feature_importances_, columns = ["Imp"], index = X_train.columns).sort_values(by = 'Imp', ascending = False)
importance_of_attributes

	Imp
no_of_employees	0.263516
prevailing_wage	0.240149
yr_of_estab	0.160512
education_of_employee_High School	0.078179
has_job_experience	0.045457
unit_of_wage_Year	0.030819
education_of_employee_Master's	0.021166
continent_Europe	0.018981
full_time_position	0.018044
requires_job_training	0.017260
education_of_employee_Doctorate	0.015964
region_of_employment_Midwest	0.015637
region_of_employment_South	0.014583
region_of_employment_Northeast	0.014157
continent_Asia	0.012366
region_of_employment_West	0.012364
continent_North America	0.009700
continent_South America	0.004952
continent_Oceania	0.002575
unit_of_wage_Week	0.002449
unit_of_wage_Month	0.001170

importances = dTree.feature_importances_
indices = np.argsort(importances)

plt.figure(figsize=(12,12))
plt.title('Feature Importances')
plt.barh(range(len(indices)), importances[indices], color='green', align='center')
plt.yticks(range(len(indices)), [feature_names[i] for i in indices])
plt.xlabel('Relative Importance')
plt.show()

As per the decision tree model, no_of_employees is the important variable for predicting

The above tree is very complex, such a tree usually overfits.

Reduce the OverFit

The deeper the tree, the more complex the model because it will have more splits and this captures more information and this is one of the causes for overfitting. Now, Lets try to limit the tree to 3.

Do we need to prune the tree?

Confusion Matrix - dTree with depth limited to 3

d_Tree_limit3 = DecisionTreeClassifier(criterion = 'gini',max_depth=3,random_state=1)
d_Tree_limit3.fit(X_train, y_train)

DecisionTreeClassifier(max_depth=3, random_state=1)

make_confusion_matrix_for_model(d_Tree_limit3, y_test)

#Using above defined function to get accuracy, recall and precision on train and test set
dTree_score_limit3=get_metrics_score(d_Tree_limit3)

Accuracy on training set :  0.7298161022650819
Accuracy on test set :  0.7243589743589743
Recall on training set :  0.926970536388819
Recall on test set :  0.929285014691479
Precision on training set :  0.7365928495197439
Precision on test set :  0.7309707241910631
f1-score on training set :  0.820888310722914
f1-score on test set :  0.8182837429926693

Lets Visualize the decision tree

plt.figure(figsize=(15,10))

tree.plot_tree(d_Tree_limit3,feature_names=feature_names,filled=True,fontsize=9,node_ids=True,class_names=True)
plt.show()

#Text Format

print(tree.export_text(d_Tree_limit3,feature_names=feature_names,show_weights=True))

|--- education_of_employee_High School <= 0.50
|   |--- has_job_experience <= 0.50
|   |   |--- unit_of_wage_Year <= 0.50
|   |   |   |--- weights: [584.00, 284.00] class: 0
|   |   |--- unit_of_wage_Year >  0.50
|   |   |   |--- weights: [2013.00, 3599.00] class: 1
|   |--- has_job_experience >  0.50
|   |   |--- education_of_employee_Master's <= 0.50
|   |   |   |--- weights: [1292.00, 3695.00] class: 1
|   |   |--- education_of_employee_Master's >  0.50
|   |   |   |--- weights: [427.00, 3524.00] class: 1
|--- education_of_employee_High School >  0.50
|   |--- continent_Asia <= 0.50
|   |   |--- continent_Europe <= 0.50
|   |   |   |--- weights: [217.00, 225.00] class: 1
|   |   |--- continent_Europe >  0.50
|   |   |   |--- weights: [231.00, 120.00] class: 0
|   |--- continent_Asia >  0.50
|   |   |--- region_of_employment_West <= 0.50
|   |   |   |--- weights: [897.00, 405.00] class: 0
|   |   |--- region_of_employment_West >  0.50
|   |   |   |--- weights: [262.00, 61.00] class: 0

# importance of features of tree building 
importance_of_the_features = pd.DataFrame(d_Tree_limit3.feature_importances_, columns = ["Imp"], index = X_train.columns).sort_values(by = 'Imp', ascending = False)
importance_of_the_features

	Imp
education_of_employee_High School	0.500464
has_job_experience	0.264116
unit_of_wage_Year	0.120021
education_of_employee_Master's	0.081335
continent_Asia	0.018963
continent_Europe	0.008847
region_of_employment_West	0.006254
yr_of_estab	0.000000
unit_of_wage_Week	0.000000
unit_of_wage_Month	0.000000
region_of_employment_South	0.000000
region_of_employment_Northeast	0.000000
region_of_employment_Midwest	0.000000
no_of_employees	0.000000
prevailing_wage	0.000000
education_of_employee_Doctorate	0.000000
requires_job_training	0.000000
continent_Oceania	0.000000
continent_North America	0.000000
full_time_position	0.000000
continent_South America	0.000000

importances_of_d_Tree_limit3 = d_Tree_limit3.feature_importances_
indices = np.argsort(importances_of_d_Tree_limit3)

plt.figure(figsize=(10,10))
plt.title('Feature Importances of d_Tree')
plt.barh(range(len(indices)), importances_of_d_Tree_limit3[indices], color='violet', align='center')
plt.yticks(range(len(indices)), [feature_names[i] for i in indices])
plt.xlabel('Relative Importance of d_Tree')
plt.show()

Observation

1. Important features changes to education_of_employee_HighSchool.
2. Recall on training set : 0.926970536388819, Recall on test set : 0.929285014691479 which shows that the model is doing better. But, accuracy is 0.72.
3. Lets see if we can further improve.

Reducing the OverFit

Using GridSearch for Hyperparameter tuning to the tree model

# Choose the type of classifier. 
estimator3 = DecisionTreeClassifier(random_state=1)

# Grid of parameters to choose from
parameters = {'max_depth': np.arange(1,10), 
              'min_samples_leaf': [1, 2, 4, 7, 9],
               "criterion": ["entropy", "gini"],
              'min_impurity_decrease': [0.001,0.01,0.1]
             }

# scoring used to compare parameter
acc_scorer_value = metrics.make_scorer(metrics.recall_score)

#  grid search
grid_obj = GridSearchCV(estimator3, parameters, scoring=acc_scorer_value,cv=5)
grid_obj = grid_obj.fit(X_train, y_train)

# Set the clf to the best combination of parameters
estimator3 = grid_obj.best_estimator_

# Fit the best algorithm to the data. 
estimator3.fit(X_train, y_train)

DecisionTreeClassifier(criterion='entropy', max_depth=1,
                       min_impurity_decrease=0.1, random_state=1)

make_confusion_matrix_for_model(estimator3,y_test)

#Using above defined function to get accuracy, recall and precision on train and test set
dTree_score_limit3=get_metrics_score(estimator3)

Accuracy on training set :  0.6679188158779995
Accuracy on test set :  0.6678440607012036
Recall on training set :  1.0
Recall on test set :  1.0
Precision on training set :  0.6679188158779995
Precision on test set :  0.6678440607012036
f1-score on training set :  0.8009008706174997
f1-score on test set :  0.8008471252647267

plt.figure(figsize=(15,10))

tree.plot_tree(estimator3,feature_names=feature_names,filled=True,fontsize=9,node_ids=True,class_names=True)
plt.show()

Bagging and Boosting Models

Building the model

1. lets build 2 ensemble models here - Bagging Classifier and Random Forest Classifier.
2. let's build these models with default parameters first and then use hyperparameter tuning to optimize the model performance.
3. The metric of interest here is recall.

Bagging Classifier

#base_estimator for bagging classifier is a decision tree by default
baggingClassifier_estimator=BaggingClassifier(random_state=1)
baggingClassifier_estimator.fit(X_train,y_train)

BaggingClassifier(random_state=1)

#Using above defined function to get accuracy, recall and precision on train and test set
bagging_estimator_score=get_metrics_score(baggingClassifier_estimator)

Accuracy on training set :  0.9850302758466024
Accuracy on test set :  0.696493982208268
Recall on training set :  0.9869050616973055
Recall on test set :  0.7737512242899118
Precision on training set :  0.9906471183013145
Precision on test set :  0.772238514173998
f1-score on training set :  0.9887725495143183
f1-score on test set :  0.7729941291585127

make_confusion_matrix_for_model(baggingClassifier_estimator,y_test)

Observation

1. Recall on training data is 0.98 and on testing data is 0.77. Lets see if there is any scope for further improvement upon tuning.

Random Forest Classifier

#Train the random forest classifier
rfClassifier_estimator=RandomForestClassifier(random_state=1)
rfClassifier_estimator.fit(X_train,y_train)

RandomForestClassifier(random_state=1)

#Using above defined function to get accuracy, recall and precision on train and test set
rf_estimator_score=get_metrics_score(rfClassifier_estimator)

Accuracy on training set :  1.0
Accuracy on test set :  0.7217425431711146
Recall on training set :  1.0
Recall on test set :  0.8366307541625857
Precision on training set :  1.0
Precision on test set :  0.7676132278936018
f1-score on training set :  1.0
f1-score on test set :  0.8006373605773738

make_confusion_matrix_for_model(rfClassifier_estimator,y_test)

Observation

1. Random Forest Model seems to over fit on training data and its 0.83 on testing data.
2. Lets see if there is any scope for the improvement.

Will tuning the hyperparameters improve the model performance?

Hyperparameter Tuning

Bagging Classifier

Some of the hyperparameters used here are max_features, n_estimators and max_samples.

# Choose the type of classifier. 
bagging_estimator_tuned = BaggingClassifier(random_state=1)

# Grid of parameters to choose from
## add from article
parameters = {
              'max_features': [0.7,0.8,0.9,1],
              'n_estimators' : [10,20,30,40,50],
              'max_samples': [0.7,0.8,0.9,1],   
             }

# Type of scoring used to compare parameter combinations
acc_scorer = metrics.make_scorer(metrics.recall_score)

# Run the grid search
grid_obj = GridSearchCV(bagging_estimator_tuned, parameters, scoring=acc_scorer,cv=5)
grid_obj = grid_obj.fit(X_train, y_train)

# Set the clf to the best combination of parameters
bagging_estimator_tuned = grid_obj.best_estimator_

# Fit the best algorithm to the data.
bagging_estimator_tuned.fit(X_train, y_train)

BaggingClassifier(max_features=0.7, max_samples=1, n_estimators=20,
                  random_state=1)

Confusion Matrix

#Using above defined function to get accuracy, recall and precision on train and test set
bagging_estimator_tuned_score=get_metrics_score(bagging_estimator_tuned)

Accuracy on training set :  0.6679188158779995
Accuracy on test set :  0.6678440607012036
Recall on training set :  1.0
Recall on test set :  1.0
Precision on training set :  0.6679188158779995
Precision on test set :  0.6678440607012036
f1-score on training set :  0.8009008706174997
f1-score on test set :  0.8008471252647267

make_confusion_matrix_for_model(bagging_estimator_tuned,y_test)

Observation

On hypertuning, the recall value on training and testing data reached 1.0.

logistic regression as the base estimator for bagging classifier:

Now, lets try use logistic regression as the base estimator.

bagging_lr=BaggingClassifier(base_estimator=LogisticRegression(solver='liblinear',random_state=1,max_iter=1000),random_state=1)
bagging_lr.fit(X_train,y_train)

BaggingClassifier(base_estimator=LogisticRegression(max_iter=1000,
                                                    random_state=1,
                                                    solver='liblinear'),
                  random_state=1)

#Using above defined function to get accuracy, recall and precision on train and test set
bagging_lr_score=get_metrics_score(bagging_lr)

Accuracy on training set :  0.6679188158779995
Accuracy on test set :  0.6678440607012036
Recall on training set :  1.0
Recall on test set :  1.0
Precision on training set :  0.6679188158779995
Precision on test set :  0.6678440607012036
f1-score on training set :  0.8009008706174997
f1-score on test set :  0.8008471252647267

make_confusion_matrix_for_model(bagging_lr,y_test)

Observation

1. Recall is 1.0 on both training and testing data.

Random Forest Classifier

Now, let's see if we can get a better model by tuning the random forest classifier. Some of the important hyperparameters available for random forest classifier are: max_sample_leaf and max_features

# Choose the type of classifier. 
rf_estimator_tuned = RandomForestClassifier(random_state=1)

# Grid of parameters to choose from
## add from article
parameters = {
    "min_samples_leaf": np.arange(5, 10),
    "max_features": np.arange(0.2, 0.7, 0.1),
             }

# Type of scoring used to compare parameter combinations
acc_scorer = metrics.make_scorer(metrics.recall_score)

# Run the grid search
grid_obj = GridSearchCV(rf_estimator_tuned, parameters, scoring=acc_scorer,cv=5)
grid_obj = grid_obj.fit(X_train, y_train)

# Set the clf to the best combination of parameters
rf_estimator_tuned = grid_obj.best_estimator_

# Fit the best algorithm to the data.
rf_estimator_tuned.fit(X_train, y_train)

RandomForestClassifier(max_features=0.2, min_samples_leaf=9, random_state=1)

#Using above defined function to get accuracy, recall and precision on train and test set
rf_estimator_tuned_score=get_metrics_score(rf_estimator_tuned)

Accuracy on training set :  0.7781453240636914
Accuracy on test set :  0.7441130298273155
Recall on training set :  0.8954083774028372
Recall on test set :  0.8748285994123408
Precision on training set :  0.7973538645537449
Precision on test set :  0.7722635310392529
f1-score on training set :  0.8435411806571507
f1-score on test set :  0.8203526818515797

make_confusion_matrix_for_model(rf_estimator_tuned,y_test)

Observation

1. We can see that random forest's performance has increased as compared to the random forest model with default parameters.
2. The test recall is slightly less.

y.head()

case_id
EZYV01    0
EZYV02    1
EZYV03    0
EZYV04    0
EZYV05    1
Name: case_status, dtype: int64

y.value_counts()

1    17018
0     8462
Name: case_status, dtype: int64

Let's try using class_weights for random forest:

The model performance is not very good. This may be due to the fact that the classes are imbalanced with 67% certified and 33% denied.

# Choose the type of classifier. 
rf_estimator_weighted = RandomForestClassifier(random_state=1)

# Grid of parameters to choose from
## add from article
parameters = {
    "class_weight": [{0: 0.3, 1: 0.7}],
    "min_samples_leaf": np.arange(5, 10),
    "max_features": np.arange(0.2, 0.7, 0.1),
}

# Type of scoring used to compare parameter combinations
acc_scorer = metrics.make_scorer(metrics.recall_score)

# Run the grid search
grid_obj = GridSearchCV(rf_estimator_weighted, parameters, scoring=acc_scorer,cv=5)
grid_obj = grid_obj.fit(X_train, y_train)

# Set the clf to the best combination of parameters
rf_estimator_weighted = grid_obj.best_estimator_

# Fit the best algorithm to the data.
rf_estimator_weighted.fit(X_train, y_train)

RandomForestClassifier(class_weight={0: 0.3, 1: 0.7}, max_features=0.2,
                       min_samples_leaf=9, random_state=1)

#Using above defined function to get accuracy, recall and precision on train and test set
rf_estimator_weighted_score=get_metrics_score(rf_estimator_weighted)

Accuracy on training set :  0.7350302758466024
Accuracy on test set :  0.7042124542124543
Recall on training set :  0.9871568874338957
Recall on test set :  0.9698334965719883
Precision on training set :  0.7200146941774322
Precision on test set :  0.7014735052422783
f1-score on training set :  0.8326842738794874
f1-score on test set :  0.8141083614239908

make_confusion_matrix_for_model(rf_estimator_weighted,y_test)

Checking the important features

importances = rf_estimator_weighted.feature_importances_
indices = np.argsort(importances)
feature_names = list(X.columns)

plt.figure(figsize=(12,12))
plt.title('Feature Importances')
plt.barh(range(len(indices)), importances[indices], color='violet', align='center')
plt.yticks(range(len(indices)), [feature_names[i] for i in indices])
plt.xlabel('Relative Importance')
plt.show()

Observation

1. The important feature is education_of_employee_highschool.
2. The model seems to be bettween after inclusing class weights. But, we also see test recall slightly less than training recall.

Boosting Model

Building the model

1. lets build 2 ensemble models here - AdaBoost Classifier, Gradient Boosting Classifier and XGBoost Classifier..
2. let's build these models with default parameters first and then use hyperparameter tuning to optimize the model performance.
3. The metric of interest here is recall.

AdaBoost Classifier

abc_1 = AdaBoostClassifier(random_state=1)
abc_1.fit(X_train,y_train)

AdaBoostClassifier(random_state=1)

#Using above defined function to get accuracy, recall and precision on train and test set
abc_score=get_metrics_score(abc_1)

Accuracy on training set :  0.7382260596546311
Accuracy on test set :  0.7352171637885924
Recall on training set :  0.8876017795685386
Recall on test set :  0.8854064642507345
Precision on training set :  0.7605005753739931
Precision on test set :  0.7585165296190636
f1-score on training set :  0.819150172367045
f1-score on test set :  0.8170643528561099

make_confusion_matrix_for_model(abc_1,y_test)

Gradient Boosting Classifier

gbc_1 = GradientBoostingClassifier(random_state=1)
gbc_1.fit(X_train,y_train)

GradientBoostingClassifier(random_state=1)

#Using above defined function to get accuracy, recall and precision on train and test set
gbc_score=get_metrics_score(gbc_1)

Accuracy on training set :  0.7564476339986544
Accuracy on test set :  0.7429356357927787
Recall on training set :  0.8800470074708302
Recall on test set :  0.8730656219392752
Precision on training set :  0.7824464512277035
Precision on test set :  0.7719085555940423
f1-score on training set :  0.8283817951959545
f1-score on test set :  0.8193767809541319

make_confusion_matrix_for_model(gbc_1,y_test)

XGBoost Classifier

xgb_1 = XGBClassifier(random_state=1,eval_metric='logloss')
xgb_1.fit(X_train,y_train)

XGBClassifier(base_score=0.5, booster='gbtree', colsample_bylevel=1,
              colsample_bynode=1, colsample_bytree=1, eval_metric='logloss',
              gamma=0, gpu_id=-1, importance_type='gain',
              interaction_constraints='', learning_rate=0.300000012,
              max_delta_step=0, max_depth=6, min_child_weight=1, missing=nan,
              monotone_constraints='()', n_estimators=100, n_jobs=8,
              num_parallel_tree=1, random_state=1, reg_alpha=0, reg_lambda=1,
              scale_pos_weight=1, subsample=1, tree_method='exact',
              validate_parameters=1, verbosity=None)

#Using above defined function to get accuracy, recall and precision on train and test set
xgb_score=get_metrics_score(xgb_1)

Accuracy on training set :  0.8384166853554609
Accuracy on test set :  0.7328623757195186
Recall on training set :  0.9283136069839671
Recall on test set :  0.8550440744368266
Precision on training set :  0.8450370596775426
Precision on test set :  0.7702488088935945
f1-score on training set :  0.8847200000000001
f1-score on test set :  0.8104344597103601

make_confusion_matrix_for_model(xgb_1,y_test)

Observation

1. AdaBoost and Gradient Boost models have better recall compared to XGBoost.

Will tuning the hyperparameters improve the model performance?

Hyperparameter Tuning

AdaBoost Classifier

Hyperparameters used here are base_estimator, n_estimators, learning_rate

# Choose the type of classifier. 
abc_tuned = AdaBoostClassifier(random_state=1)

# Grid of parameters to choose from
## add from article
parameters = {
    #Let's try different max_depth for base_estimator
    "base_estimator":[DecisionTreeClassifier(max_depth=1, random_state=1),DecisionTreeClassifier(max_depth=2, random_state=1),DecisionTreeClassifier(max_depth=3, random_state=1),DecisionTreeClassifier(max_depth=4, random_state=1)],
    "n_estimators": np.arange(10,110,20),
    "learning_rate":np.arange(0.1,2,0.1)
}

# Type of scoring used to compare parameter combinations
acc_scorer = metrics.make_scorer(metrics.recall_score)

# Run the grid search
grid_obj = GridSearchCV(abc_tuned, parameters, scoring=acc_scorer,cv=5)
grid_obj = grid_obj.fit(X_train, y_train)

# Set the clf to the best combination of parameters
abc_tuned = grid_obj.best_estimator_

# Fit the best algorithm to the data.
abc_tuned.fit(X_train, y_train)

AdaBoostClassifier(base_estimator=DecisionTreeClassifier(max_depth=1,
                                                         random_state=1),
                   learning_rate=0.2, n_estimators=10, random_state=1)

#Using above defined function to get accuracy, recall and precision on train and test set
abc_tuned_score=get_metrics_score(abc_tuned)

Accuracy on training set :  0.6919152276295133
Accuracy on test set :  0.6890371533228676
Recall on training set :  0.9709560983799211
Recall on test set :  0.9704211557296768
Precision on training set :  0.6919717635798038
Precision on test set :  0.6899721448467967
f1-score on training set :  0.8080617555625416
f1-score on test set :  0.8065120065120066

make_confusion_matrix_for_model(abc_tuned,y_test)

importances = abc_tuned.feature_importances_
indices = np.argsort(importances)
feature_names = list(X.columns)

plt.figure(figsize=(12,12))
plt.title('Feature Importances')
plt.barh(range(len(indices)), importances[indices], color='violet', align='center')
plt.yticks(range(len(indices)), [feature_names[i] for i in indices])
plt.xlabel('Relative Importance')
plt.show()

1. education_of_employee_highschool is the important feature as per tunes AdaBoost.

Gradient Boosting Classifier

Let's try using AdaBoost classifier as the estimator for initial predictions

gbc_init = GradientBoostingClassifier(init=AdaBoostClassifier(random_state=1),random_state=1)
gbc_init.fit(X_train,y_train)

GradientBoostingClassifier(init=AdaBoostClassifier(random_state=1),
                           random_state=1)

#Using above defined function to get accuracy, recall and precision on train and test set
gbc_init_score=get_metrics_score(gbc_init)

Accuracy on training set :  0.7573446961202063
Accuracy on test set :  0.7438513867085296
Recall on training set :  0.8818097876269622
Recall on test set :  0.875024485798237
Precision on training set :  0.7824953445065177
Precision on test set :  0.7719025401762571
f1-score on training set :  0.8291893598547635
f1-score on test set :  0.8202350348879912

Gradient Boosting Tuned

As compared to the default parameters, the metrics seems to be same

# Choose the type of classifier. 
gbc_tuned = GradientBoostingClassifier(init=AdaBoostClassifier(random_state=1),random_state=1)

# Grid of parameters to choose from
## add from article
parameters = {
    "n_estimators": [50,100,150,200,250],
    "subsample":[0.8,0.9,1],
    "max_features":[0.7,0.8,0.9,1]
}

# Type of scoring used to compare parameter combinations
acc_scorer = metrics.make_scorer(metrics.recall_score)

# Run the grid search
grid_obj = GridSearchCV(gbc_tuned, parameters, scoring=acc_scorer,cv=5)
grid_obj = grid_obj.fit(X_train, y_train)

# Set the clf to the best combination of parameters
gbc_tuned = grid_obj.best_estimator_

# Fit the best algorithm to the data.
gbc_tuned.fit(X_train, y_train)

GradientBoostingClassifier(init=AdaBoostClassifier(random_state=1),
                           max_features=1, n_estimators=50, random_state=1,
                           subsample=0.8)

#Using above defined function to get accuracy, recall and precision on train and test set
gbc_tuned_score=get_metrics_score(gbc_tuned)

Accuracy on training set :  0.7205090827539807
Accuracy on test set :  0.7091836734693877
Recall on training set :  0.9456056408964997
Recall on test set :  0.9475024485798237
Precision on training set :  0.7220228175874888
Precision on test set :  0.7121613663133097
f1-score on training set :  0.8188260948573505
f1-score on test set :  0.8131461713036899

make_confusion_matrix_for_model(gbc_tuned,y_test)

importances = gbc_tuned.feature_importances_
indices = np.argsort(importances)
feature_names = list(X.columns)

plt.figure(figsize=(12,12))
plt.title('Feature Importances')
plt.barh(range(len(indices)), importances[indices], color='violet', align='center')
plt.yticks(range(len(indices)), [feature_names[i] for i in indices])
plt.xlabel('Relative Importance')
plt.show()

Observation

1. The recall value has increased compared to the previous models.
2. education_of_employee_HighSchool is the important feature.

XGBoost Classifier

# Choose the type of classifier. 
xgb_tuned = XGBClassifier(random_state=1,eval_metric='logloss')

# Grid of parameters to choose from
## add from
parameters = {
    "n_estimators": np.arange(10,100,20),
    "scale_pos_weight":[0,1,2,5],
    "subsample":[0.5,0.7,0.9,1],
    "learning_rate":[0.01,0.1,0.2],
    "colsample_bytree":[0.5,0.7,0.9],
    "colsample_bylevel":[0.5,0.7,0.9]
}

# Type of scoring used to compare parameter combinations
acc_scorer = metrics.make_scorer(metrics.recall_score)

# Run the grid search
grid_obj = GridSearchCV(xgb_tuned, parameters,scoring=acc_scorer,cv=5)
grid_obj = grid_obj.fit(X_train, y_train)

# Set the clf to the best combination of parameters
xgb_tuned = grid_obj.best_estimator_

# Fit the best algorithm to the data.
xgb_tuned.fit(X_train, y_train)

XGBClassifier(base_score=0.5, booster='gbtree', colsample_bylevel=0.5,
              colsample_bynode=1, colsample_bytree=0.5, eval_metric='logloss',
              gamma=0, gpu_id=-1, importance_type='gain',
              interaction_constraints='', learning_rate=0.01, max_delta_step=0,
              max_depth=6, min_child_weight=1, missing=nan,
              monotone_constraints='()', n_estimators=10, n_jobs=8,
              num_parallel_tree=1, random_state=1, reg_alpha=0, reg_lambda=1,
              scale_pos_weight=5, subsample=0.5, tree_method='exact',
              validate_parameters=1, verbosity=None)

#Using above defined function to get accuracy, recall and precision on train and test set
xgb_tuned_score=get_metrics_score(xgb_tuned)

Accuracy on training set :  0.6679188158779995
Accuracy on test set :  0.6678440607012036
Recall on training set :  1.0
Recall on test set :  1.0
Precision on training set :  0.6679188158779995
Precision on test set :  0.6678440607012036
f1-score on training set :  0.8009008706174997
f1-score on test set :  0.8008471252647267

make_confusion_matrix_for_model(xgb_tuned,y_test)

importances = xgb_tuned.feature_importances_
indices = np.argsort(importances)
feature_names = list(X.columns)

plt.figure(figsize=(12,12))
plt.title('Feature Importances')
plt.barh(range(len(indices)), importances[indices], color='violet', align='center')
plt.yticks(range(len(indices)), [feature_names[i] for i in indices])
plt.xlabel('Relative Importance')
plt.show()

Observation

1. Important Feature is education_of_employee_High School.
2. Though recall is 1.0, the accuracy seems to be very less.

Model Improvement Boosting

Stacking Model

estimators_1=[('AdaBoost', abc_1),('Gradiant Boost', gbc_1)]
final_estimator_1=RandomForestClassifier(random_state=1)

stacking_estimator_1=StackingClassifier(estimators=estimators_1, final_estimator=final_estimator_1,cv=5)
stacking_estimator_1.fit(X_train,y_train)

StackingClassifier(cv=5,
                   estimators=[('AdaBoost', AdaBoostClassifier(random_state=1)),
                               ('Gradiant Boost',
                                GradientBoostingClassifier(random_state=1))],
                   final_estimator=RandomForestClassifier(random_state=1))

stacking_Score_1=get_metrics_score(stacking_estimator_1)

Accuracy on training set :  0.716640502354788
Accuracy on test set :  0.7008110936682366
Recall on training set :  0.8270796608746748
Recall on test set :  0.8154750244857982
Precision on training set :  0.7669494823694247
Precision on test set :  0.7558097312999273
f1-score on training set :  0.7958804523424878
f1-score on test set :  0.7845095637425799

Observations

1. All the metrics obtained are pretty decent value.
2. Recall is lesser compared to the value obtained in the previous model. Lets, compare all the models.

Model Performance Comparison and Conclusions

# defining list of models
models = [dTree,d_Tree_limit3,estimator3, baggingClassifier_estimator,bagging_estimator_tuned,bagging_lr,rfClassifier_estimator,rf_estimator_tuned,
          rf_estimator_weighted, abc_1, abc_tuned, gbc_1, gbc_init, gbc_tuned, xgb_1, xgb_tuned, stacking_estimator_1]



# defining empty lists to add train and test results
acc_train = []
acc_test = []
recall_train = []
recall_test = []
precision_train = []
precision_test = []
fscore_train = []
fscore_test =  []


# looping through all the models to get the accuracy, precall and precision scores
for model in models:
    j = get_metrics_score(model,False)
    acc_train.append(np.round(j[0],2))
    acc_test.append(np.round(j[1],2))
    recall_train.append(np.round(j[2],2))
    recall_test.append(np.round(j[3],2))
    precision_train.append(np.round(j[4],2))
    precision_test.append(np.round(j[5],2))
    fscore_train.append(np.round(j[6],2))
    fscore_test.append(np.round(j[7],2))

comparison_frame = pd.DataFrame({'Model':['Decision Tree', 'Decision tree with max_depth =3', 'Decision Tree Hypertuned', 'baggingClassifier_estimator',
                                          'bagging_estimator_tuned', 'bagging_logistic_regression',' Random Forest Classifier','Random Forest Classifier - tuned', 'Random Forest Classifier -weighted','AdaBoost with default paramters','AdaBoost Tuned', 'Gradient Boosting with default parameters','Gradient Boosting with init=AdaBoost',
                                          'Gradient Boosting Tuned','XGBoost with default parameters','XGBoost Tuned','Stacking Model'], 
                                          'Train_Accuracy': acc_train,'Test_Accuracy': acc_test,
                                          'Train_Recall':recall_train,'Test_Recall':recall_test,
                                          'Train_Precision':precision_train,'Test_Precision':precision_test,
                                          'Train_f1score':fscore_train,'Test_f1score':fscore_test,
                                        }) 
comparison_frame

	Model	Train_Accuracy	Test_Accuracy	Train_Recall	Test_Recall	Train_Precision	Test_Precision	Train_f1score	Test_f1score
0	Decision Tree	1.00	0.66	1.00	0.74	1.00	0.75	1.00	0.74
1	Decision tree with max_depth =3	0.73	0.72	0.93	0.93	0.74	0.73	0.82	0.82
2	Decision Tree Hypertuned	0.67	0.67	1.00	1.00	0.67	0.67	0.80	0.80
3	baggingClassifier_estimator	0.99	0.70	0.99	0.77	0.99	0.77	0.99	0.77
4	bagging_estimator_tuned	0.67	0.67	1.00	1.00	0.67	0.67	0.80	0.80
5	bagging_logistic_regression	0.67	0.67	1.00	1.00	0.67	0.67	0.80	0.80
6	Random Forest Classifier	1.00	0.72	1.00	0.84	1.00	0.77	1.00	0.80
7	Random Forest Classifier - tuned	0.78	0.74	0.90	0.87	0.80	0.77	0.84	0.82
8	Random Forest Classifier -weighted	0.74	0.70	0.99	0.97	0.72	0.70	0.83	0.81
9	AdaBoost with default paramters	0.74	0.74	0.89	0.89	0.76	0.76	0.82	0.82
10	AdaBoost Tuned	0.69	0.69	0.97	0.97	0.69	0.69	0.81	0.81
11	Gradient Boosting with default parameters	0.76	0.74	0.88	0.87	0.78	0.77	0.83	0.82
12	Gradient Boosting with init=AdaBoost	0.76	0.74	0.88	0.88	0.78	0.77	0.83	0.82
13	Gradient Boosting Tuned	0.72	0.71	0.95	0.95	0.72	0.71	0.82	0.81
14	XGBoost with default parameters	0.84	0.73	0.93	0.86	0.85	0.77	0.88	0.81
15	XGBoost Tuned	0.67	0.67	1.00	1.00	0.67	0.67	0.80	0.80
16	Stacking Model	0.72	0.70	0.83	0.82	0.77	0.76	0.80	0.78

Observation

1. According to the Recall definition, high recall means less false negatives. Lower chances of rejected cases as certified. Recall should be maximized, the greater the Recall higher the chances of identifying.
2. Here, the highest recall has been obtained for Decision Tree Hypertuned, bagging_logistic_regression, bagging_estimator_tuned and XGBoost Tuned. But, the accuracy of these models are very low (0.67)
3. Coming to the second best models better recall value and accuracy are - Random Forest Classifier -weighted and Gradient Boosting Tuned
4. The differnce between recall values of Random Forest Classifier -weighted and Gradient Boosting Tuned models is 2%. The recall value of Random Forest Classifier -weighted is slightly more.
5. The accuracy of Gradient Boosting Tuned models is 1% more than Random Forest Classifier -weighted.
6. But, the differnces between respective training vs testing recall values is less in Gradient Boosting Tuned model.
7. Hence, Gradient Boosting Tuned model is preferred.

Actionable Insights and Recommendations

From the data analysis, we can see definite patterns.

1. Most applications and certifications are from Asia
2. Close to 70% have Masters or Bachelors.
3. People from Europe have a low certification rate.
4. People from South America, North America and Asia have high certification rate.
5. Certification is highest for people with Doctorate or Masters degree.
6. Certification is lowest for people with High school degree.
7. Rejection rate is high in people with no job experience.
8. All weekly wage workers are certified.
9. Employees of Northeast has higher percentage of certification while employees of Midwest have the highest percentage of rejection. Hence Education of the candidates (High school / Masters / Doctorate), Job experience, Prevailing wage, Unit of wage and region of employment play a key role in the certification rate.

OFLC can save a lot of resources and time by pre-filtering/ pre-sorting the candidates based on the model. They can start processing the applications where the candidates have high level of education, job experience, prevailing wage and being employed in regions like the North East. They could also raise the minimum requirements for prevailing wage, Education, job experience so that they get only high quality candidates who are more likely to be certified. Since the model selected has high recall and a decent accuracy, the model predictions on positive certifications will help reduce resource wastage while keeping the opportunity cost of losing good candidates to the minimum.